Plan a geographical investigation by setting a question and hypothesis, choosing primary and secondary data, and selecting a sensible sampling method

What this dot point is asking

This skill asks you to plan a geographical investigation: to turn a topic into a clear question and a testable hypothesis, to decide what data you need and whether it is primary or secondary, and to choose a sensible way of sampling and recording it. The data-response section often asks you to suggest or evaluate fieldwork methods, so you need to understand the process, not just memorise it. The central idea is that good data starts with good planning: a sharp question and the right method give answers you can trust.

The answer

Asking a geographical question and writing a hypothesis

Every investigation starts with a focused geographical question tied to a place and something you can measure, for example "Is traffic heavier near the school in the morning than at midday?" From the question you write a hypothesis, a clear statement you can test with data, such as "There are more vehicles passing the school gate between 7 and 8 am than between 12 and 1 pm." A good hypothesis is specific, measurable and could turn out to be true or false.

Primary and secondary data

Primary data is data you collect yourself in the field: counts, measurements, photographs and questionnaires. It is up to date and tailored to your question. Secondary data is data someone else has already collected and published: government statistics, maps, weather records and reports. It saves time and can cover larger areas or longer periods. A strong investigation usually uses both, with secondary data giving context and primary data answering the exact question.

Choosing a sampling method

It is rarely possible to measure everything, so you take a sample. The three common methods are:

• Random sampling: every item or place has an equal chance of being chosen, for example using random numbers to pick survey points. It avoids bias but can miss areas by chance.
• Systematic sampling: you choose at regular intervals, for example every tenth person or a measurement every 50 metres. It is simple and gives even coverage.
• Stratified sampling: you split the group into categories (for example age groups or land-use types) and sample each in proportion. It makes sure each group is represented.

The right method depends on the aim; you should always say why you chose it.

Collecting reliable data

Data is only useful if it is collected carefully. Use a prepared recording sheet such as a tally chart or a table so nothing is forgotten. Keep conditions the same when comparing (same length of time, same spot) so the comparison is fair. Repeat readings where possible to reduce the effect of a one-off result, and note the date, time and place so the data can be understood and repeated later.

Examples in context

Example 1. A traffic study near a Singapore school. Students testing whether traffic peaks at the morning bell can count vehicles passing the gate in fixed five-minute slots (systematic sampling) at 7.30 am and again at noon, recording on a tally sheet. Secondary data, such as the school start time and a road map, gives context. Keeping the slot length and location the same makes the comparison fair.

Example 2. A questionnaire on visitor origins at a tourist site. To find out where visitors to a heritage site come from, a group might use a short questionnaire and stratified sampling, asking a set number of visitors in each age group. This primary data, combined with secondary tourism-board statistics, gives a fuller picture than either source alone.

Try this

Q1. Write a testable hypothesis for the question "Is the park busier on weekends than on weekdays?" [2 marks]

• Cue. For example: more people enter the park on Saturday than on a Tuesday, counted over the same one-hour period.

Q2. State one advantage of using systematic sampling. [1 mark]

• Cue. It is simple to carry out and gives even, regular coverage, for example a measurement every 50 metres.

Q3. Explain one way to make a people count more reliable. [2 marks]

• Cue. Repeat the count at the same spot for the same length of time across the day, so a single busy or quiet moment does not distort the result.