Geographic inquiry and sampling
Introduit| English | Sens |
|---|---|
| geographic inquiry | an investigation that asks a geographical question, collects data, analyses it and reaches a conclusion |
| hypothesis/haɪˈpɒθəsɪs/ | A proposed explanation or prediction that can be tested using evidence. |
| research question/rɪˈsɜːtʃ ˈkwestʃn/ | The specific question an investigation is designed to answer. |
| sampling/ˈsæmplɪŋ/ | échantillonnage |
| random sampling/ˈrændəm ˈsæmplɪŋ/ | Sampling in which every point or person has an equal chance of being chosen. |
| systematic sampling/ˌsɪstəˈmætɪk ˈsæmplɪŋ/ | Sampling at regular intervals, such as every tenth metre or every fifth person. |
| stratified sampling/ˈstrætɪfaɪd ˈsæmplɪŋ/ | Sampling from each group in proportion to its share of the whole population or area. |
| fieldwork/ˈfiːldwɜːk/ | travail de terrain |
A map of deaths around a pump, 1854
- In 1854 cholera killed over 500 people in ten days in Soho, London. The doctor John Snow marked each death on a street map and found them clustered around one water pump on Broad Street. He persuaded officials to remove the pump handle.
- Snow collected data systematically in the field, mapped it and tested an idea against it. That is geographic inquiry 地理探究.
What did John Snow's 1854 map show?
Difficulté 1/5 · 100 pièces + un petit bonus lors de la première réussite aujourd'hui
Systematic mapping revealed the pattern.
Question and hypothesis
- Start with a focused research question 研究问题: "How does pedestrian flow change with distance from the shopping centre?"
- A hypothesis 假设 is a testable prediction: "Pedestrian counts decrease with distance from the shopping centre."
Which is a testable hypothesis?
Difficulté 1/5 · 100 pièces + un petit bonus lors de la première réussite aujourd'hui
A hypothesis is a testable prediction.
Sampling
- Random sampling 随机抽样: every site has an equal chance. Systematic sampling 系统抽样: sites at regular intervals, such as every 100 m. Stratified sampling 分层抽样: samples in proportion to sub-groups, such as land uses.
- Systematic sampling along a line from the centre suits a question about distance.

Match each sampling method to its description.
Difficulté 2/5 · 140 pièces + un petit bonus lors de la première réussite aujourd'hui
Choose the method that fits the question.
Worked example: testing the hypothesis
Mean five-minute pedestrian counts (three repeats each): site 1 (0 m) 84, site 2 (100 m) 61, site 3 (200 m) 47, site 4 (300 m) 22, site 5 (400 m) 25.
- Pattern: counts fall from 84 to 22 over 300 m, then rise slightly at 400 m.
- Spearman's rank: distance ranks 1–5 against count ranks (lowest count = 1) of 5, 4, 3, 1, 2 give $\sum d^2 = 16 + 4 + 0 + 9 + 9 = 38$, so $r_s = 1 - \frac{6 \times 38}{5(25 - 1)} = -0.9$: a strong negative correlation — the further from the centre, the fewer pedestrians.
- Anomaly: site 5 is next to a bus stop, which explains the small rise. With only five sites, the result is a strong pattern, not proof.
In the worked example, what is Spearman's rank correlation coefficient?
Difficulté 2/5 · 140 pièces + un petit bonus lors de la première réussite aujourd'hui
1 − (6 × 38) ÷ (5 × 24) = −0.9, a strong negative correlation.
Why did the count rise slightly at site 5?
Difficulté 1/5 · 100 pièces + un petit bonus lors de la première réussite aujourd'hui
Explain anomalies with local factors.
Fair and safe fieldwork
- Keep conditions comparable: the same count length, similar times of day, the same weather where possible, repeats at each site.
- Work in pairs, follow a risk assessment, and do not record personal information about people counted.
Which keep a pedestrian count fair?
Difficulté 2/5 · 140 pièces + un petit bonus lors de la première réussite aujourd'hui
Comparable conditions make data comparable.
The internal assessment
- The fieldwork report (up to 2500 words) states the question and hypothesis, explains methods, presents and analyses data, and concludes and evaluates.
- Evaluation names specific weaknesses — five sites, one day, an anomaly — and specific improvements.
A correlation in fieldwork data does not prove the cause. Pedestrian counts may fall with distance because of shops, or because of a bus stop or a school. Name other factors and check them.
A strong correlation in fieldwork data proves the cause.
Difficulté 1/5 · 100 pièces + un petit bonus lors de la première réussite aujourd'hui
Name and check other factors.
Geographic inquiry starts with a focused question and testable hypothesis. Choose random, systematic or stratified sampling to fit the question, keep conditions fair and safe, analyse with tools like Spearman's rank, explain anomalies, and evaluate specific limits.
Design a fieldwork investigation with a question, hypothesis and sampling method.
Question, hypothesis, sampling, analysis.