Fieldwork That Answers a Question
Plan fieldwork around a question that observations can test. Define what to measure, choose sites and times, justify the sample, and identify possible biases while keeping the work safe.
What a learner can do afterwards
- Turn a vague interest into a question that observations can test
- Choose a sampling method and justify it against the alternatives for that site
- Identify two things about your method that could bias the result, and say how you would reduce each
1 · Read
We now turn from world trade to local fieldwork. Earlier fieldwork used maps, tallies and measurements. Plan a test before collecting more data. To study litter, count equal-sized plots near food shops and farther away at noon. Decide what counts as one item. Similar counts would not support a prediction of more litter near shops. Questions about people's views need a survey instead. A sample without a clear difference does not prove there is no pattern elsewhere.
A sample covers part of a defined study area. For a simple random sample, list eligible spots and use a chance process giving each an equal chance. Systematic sampling uses fixed gaps, preferably after a random start. Repeated features can match the gaps. Stratified sampling divides the area into zones and samples within each. A whole-site estimate must also account for the zones' different sizes.
Justify the method using the question and site. To compare dunes, a car park and a pier, sample within every permitted zone; a small random sample might miss one. Along a path, fixed gaps spread stations along its length, while a small random sample may leave uneven gaps. State a limitation too: repeating features may line up with fixed stations. No method is best for every enquiry.
Name possible biases and a change for each. Dry-day visitor counts may miss wet-weather use; repeat at planned times in varied weather when conditions are safe. A leading question suggests an answer and may influence replies. Replace it with neutral wording. Pilot the questionnaire by testing it on a small group, then revise it. A pilot helps find problems but does not remove every bias.
Safety and permission come before coverage. Do not enter unsafe or restricted ground to complete a sample. Define an approved study area, then choose stations across it rather than only beside the easiest path. Record excluded places. If they differ from the sampled ground, limit conclusions to the area you could study instead of claiming to describe the whole site.
Define the question and study area, justify the sample, keep the work safe, and report possible biases and limits.
Fieldwork planning teaching, questions and flashcards written for this dataset.
These fieldwork readings and flashcards were written for this dataset.
More to read
Write the comparison before the visit
A class wants to investigate litter around a shopping square. Before visiting, the class must decide what one litter item means and how much ground each count will cover. Counting a large area near a shop and a tiny area elsewhere would not give a fair comparison of amounts per equal area.
Plan equal-sized plots near shops and farther away. Use the same counting rules and time. Write down what result would support the prediction and what result would not. Similar counts would leave the prediction unsupported by these observations. Report that finding with the method. It would not establish that litter patterns are absent from every shopping square.
Three ways to place stations
First mark the approved study area and the spots that could be sampled. A simple random draw uses chance to select from a complete list of eligible spots, giving each the same chance. The selected spots may still be unevenly spread.
A systematic plan uses a fixed interval along a route. Choosing the start by chance helps avoid picking a favoured starting location. Check whether regular features might coincide with the interval.
A stratified plan divides the area into distinct zones and collects observations within every zone. This is useful for comparing zones. Estimating a result for the whole area also requires accounting for how much of the area each zone occupies. The question determines which coverage matters.
Give a reason another reader can check
Calling a method stratified or systematic does not explain why you chose it. Describe the problem it solves at your site. To compare three habitat zones, a plan that samples within all three provides observations for each comparison. A small draw from one whole-site list might omit a zone.
Then explain a cost or limitation. Dividing a site into zones requires useful zone boundaries. Fixed stations along a route can coincide with regular features. A random draw can leave uneven gaps. Keep a record of the planned stations and any changes. A reader should be able to compare your reasons with what you actually did.
Improve coverage within safe limits
Suppose visitor counts were collected only in dry weather. Those records do not directly describe wet-weather use. Additional visits at planned times in other conditions may improve coverage, but only when the work is safe and permitted. Do not enter hazardous ground to complete a sampling plan.
Record any access exclusions and explain how they limit the conclusion. Within approved ground, selecting only the easiest stations may still miss different conditions elsewhere.
Question wording needs checking too. A leading question can suggest a preferred answer. Use neutral wording, try the draft with a small group and revise unclear or leading questions. This pilot can reveal problems; it does not guarantee a bias-free survey.
2 · Watch
Take it off screen
Where it sits
Optional connections
This opens up
Where this leads
Jobs that lean on this skill. Follow one to see everything it is built on.
8 questions wait behind this lesson, each with its answer explained. Every answer feeds the sky: stars light as they are learned, and dim when it is time to come back.