Skip to content
Tutorials Center
My learning

Topics

Lesson 35 of 38 ยท Projects and Practice

A Geography Enquiry: Planning Your Own Investigation

Learn the steps of a geographical enquiry, how to choose a question, collect and sample data, present results and test a link with the Spearman rank correlation.

Geography is not just facts to memorize. It is a way of asking questions about places and finding evidence to answer them. Many courses include an independent investigation, often with fieldwork: collecting data outdoors, in a street, a river, a park or a town center. Even if you cannot go outside, you can run the same process with data from maps, statistics and surveys.

In this project lesson you will learn the steps of an enquiry, how to plan data collection, and how to test whether two things are related using a simple statistic.

The enquiry cycle

flowchart LR
    A["1 Question"] --> B["2 Plan"]
    B --> C["3 Collect data"]
    C --> D["4 Present"]
    D --> E["5 Analyze"]
    E --> F["6 Conclude"]
    F --> G["7 Evaluate"]

The seven stages of a geographical enquiry.

1. Choose a question and hypothesis

A good enquiry question is focused, testable and answerable with data you can get. Compare these:

  • Weak: "What is a city like?" (too broad)
  • Better: "How does the number of pedestrians change with distance from the town center?"
  • Better: "Does a river's width increase from source to mouth?"

A hypothesis is a clear statement you can test, such as "Pedestrian numbers decrease as distance from the town center increases." A null hypothesis says there is no relationship. Your data will either support the hypothesis or fail to.

2. Plan

Decide:

  • What to measure: the variables. Here, distance from the center (the independent variable) and pedestrian count (the dependent variable).
  • Where and when: same time of day and week for every site, so the comparison is fair.
  • How: which tools (tape measure, tally sheet, flow meter, questionnaire, app).
  • Risk assessment: hazards such as traffic, deep water or weather, and how to reduce them. Never do fieldwork alone or without permission where needed.
  • Ethics: ask permission, keep people's answers anonymous, and do not photograph individuals without consent.

3. Collect data

Primary data is information you collect yourself (counts, measurements, interviews). Secondary data is information collected by others (census data, maps, weather records, satellite images). Both are useful; secondary data can provide context and a check on your own results.

Quantitative data is numbers; qualitative data is descriptions, opinions and photos. A good enquiry often uses both.

Sampling

You rarely have time to measure everything, so you take a sample: a smaller group that represents the whole.

Method How Good for Risk
Random Each site or person has an equal chance of selection Avoiding bias May miss important areas by chance
Systematic Take every nth item, for example a site every 100 m Easy and even coverage Can match a hidden pattern by accident
Stratified Divide into groups (strata) and sample each in proportion Making sure each part is represented Needs prior knowledge of the groups

A larger sample gives more reliable results. A common rule of thumb for fieldwork is to take at least 10 to 30 data points if possible, but the best number depends on the question.

Python Choose sampling sites

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Presenting data

Choose the display that suits the data:

  • Bar chart: categories (for example, land use).
  • Histogram: counts in ranges, such as pebble sizes.
  • Line graph: change over time or distance.
  • Scatter graph: the relationship between two measurements; add a line of best fit.
  • Pie chart: shares of a whole, only for a few categories.
  • Annotated map or field sketch: shows where things are.

Always label axes with units and add a title.

Python Scatter graph of a transect

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Analyzing: is there a relationship?

A graph suggests a relationship, but how strong is it? For ranked data, geographers often use the Spearman rank correlation coefficient (rs). It runs from โˆ’1 to +1:

  • +1: a perfect positive relationship (as one rises, so does the other).
  • โˆ’1: a perfect negative relationship (as one rises, the other falls).
  • 0: no relationship.

Steps: rank each variable from 1 to n, find the difference d between each pair of ranks, square them, add them up and use

Text
rs = 1 โˆ’ (6 ร— ฮฃdยฒ) รท (n ร— (nยฒ โˆ’ 1))

The formula is simple when no values tie. (If values tie, use the average of their ranks; the simple formula is then a close approximation.)

Python Spearman rank calculator

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

A result near โˆ’1 supports the hypothesis that pedestrian numbers decrease with distance. Remember: a relationship (correlation) does not prove that one thing causes the other.

Concluding and evaluating

  • Conclusion: answer your question using your results. State clearly whether the hypothesis was supported, quote numbers, and link to geographical theory (for example, the concentric zone model).
  • Evaluation: discuss reliability. Were there anomalies (odd results) and why? Was the sample big enough? Did weather or events on the day affect results? How could methods be improved, for example more sites, more days, or a better tool?
  • Limitations: results from one town on one day cannot be applied everywhere.

Write down your data in a table on the day, with the date, time and weather. If you cannot remember where a number came from, you cannot use it.

Planning checklist

Use this checklist to make sure your project is ready.

JS Enquiry plan checker

Press Run (or Ctrl + Enter) to see the console output here.

No output.

Two items show as missing: ethics and secondary data. Fill them in to complete the plan.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. Which is the best enquiry question?

    A good question is focused, measurable and answerable with data you can collect.
  2. 2. A systematic sample takes...

    Systematic sampling picks items at a regular interval, for example every 4th site.
  3. 3. A Spearman rank result of โˆ’0.9 suggests...

    Values near โˆ’1 mean that as one variable increases, the other tends to decrease.
  4. 4. Which is secondary data?

    Secondary data is collected by someone else; primary data is gathered by you.

Summary

  • An enquiry moves from question to plan, data collection, presentation, analysis, conclusion and evaluation.
  • Good questions are focused and testable; hypotheses state what you expect.
  • Use primary and secondary data, and sample fairly (random, systematic or stratified).
  • Present data with labelled graphs and maps; test relationships with a statistic such as Spearman's rank.
  • Correlation is not causation; always evaluate your method.

Next: Project: Build a Country Profile