Introduction: Why Sampling and Data Integrity Matter
Imagine you wanted to find out the average size of every pebble on a beach or the opinion of every single person living in London. It would be impossible! You would run out of time, money, and energy. In Geography, we use sampling to take a "snapshot" of the real world. However, if that snapshot is blurry (measurement error) or if we zoom in on only the bits we like (misuse of data), our conclusions will be wrong. This chapter teaches you how to be a "data detective" to ensure your fieldwork and research are accurate and ethical.
1. Sampling Strategies
Sampling is the process of collecting data from a small group (the sample) to represent a larger group (the population). For your AS Level fieldwork in Paper 1 or Paper 2, you must choose the right strategy to avoid bias.
A. Random Sampling
In random sampling, every member of the population has an equal chance of being selected. You might use a random number generator to pick coordinates on a map or use a random number table to select pebbles.
- Pro: It is completely objective and removes human bias.
- Con: You might accidentally miss certain areas or groups if the random points "clump" together.
B. Systematic Sampling
This involves collecting data at regular, fixed intervals. For example, measuring longshore drift every \(10\) metres along a beach, or interviewing every \(5\)th person who passes a shop.
- Pro: It provides good "spatial coverage," ensuring you look at the whole area.
- Con: It can be biased if there is a pattern in the landscape that matches your interval (e.g., if you measure every \(10\) metres and there is a sea groynes every \(10\) metres).
C. Stratified Sampling
This is used when your population has distinct sub-groups (strata). If you are studying a town where \(70\%\) of people are over 60 and \(30\%\) are under 60, your sample of \(100\) people should include \(70\) seniors and \(30\) young people to be fair.
- Pro: It ensures that small but important groups are not ignored.
- Con: You need prior knowledge or secondary data (like Census data) to know the proportions of the groups first.
Quick Tip: If a question asks how to improve a sample, usually the answer is "increase the sample size" (represented as \(n\)). The larger the \(n\), the more reliable the data!
2. Measurement Error: When Things Go Wrong
Even with a perfect sampling plan, mistakes happen during data collection. We call these measurement errors. Understanding these helps you "evaluate" your fieldwork in the exam.
Human Error
These are mistakes made by the researcher. Example: Misreading a clinometer while measuring a slope, or writing down "10.5m" instead of "1.05m." Parallax error is a common one—this is when you read a scale from a weird angle, making the measurement look higher or lower than it actually is.
Instrument Error
Sometimes the equipment is the problem. Example: A digital anemometer (wind speed measurer) might have low batteries and give slow readings, or a tape measure might have stretched over years of use. This is often called a systematic error because it affects every single measurement by the same amount.
Sampling Error
This happens when the sample you chose doesn't actually represent the whole population. Example: If you only measure pebbles at the top of a beach (where the big ones are) and conclude the whole beach is made of boulders, you have a sampling error.
Key Takeaway: Always "calibrate" your equipment (check it works against a known standard) and take multiple readings to find an average to reduce the impact of random errors.
3. Misuse of Data and Ethics
Geography isn't just about numbers; it's about people and places. Data can be misused—intentionally or accidentally—to tell a misleading story.
Bias and Manipulation
Data is misused when a researcher "cherry-picks" results. Example: A property developer might only show data about the new jobs created by a regeneration project (Topic 4A) but hide data about the local people who were forced to move out due to rising rents.
Ethical Implications
When collecting data about human communities (especially for Diverse Places or Regenerating Places), you must follow ethical guidelines:
- Informed Consent: People must agree to be interviewed and know how their data will be used.
- Anonymity: You should not record names or house numbers that could identify individuals.
- Sensitivity: Some topics (like poverty or migration) are personal. Asking questions in a pushy or judgmental way is an ethical misuse of the research process.
Socio-political Implications
Data can be used to influence politics. For example, if a government only collects data on "economic growth" but ignores "environmental quality," they might justify building a factory that pollutes a local river. As a geographer, you must ask: Who collected this data, and what is their goal?
Don't forget: In Paper 2, you may be asked to evaluate the "lived experience" of a place. Quantitative data (numbers) might show a place is "improving," but qualitative data (interviews) might show that residents feel unhappy or excluded.
Quick Review: Avoiding Common Pitfalls
Common Mistake: Students often think "Random Sampling" means "just picking whatever is nearby." The Correction: That is actually called Opportunity Sampling and it's very biased! True Random Sampling requires a specific system, like a random number generator.
Fieldwork Link: In your exam, if you are asked to "Assess the reliability of your data," mention:
- Your sampling strategy (and if it was large enough).
- Potential measurement errors (human or instrument).
- Whether you remained objective and ethical during collection.
Note: For more information on how to display this data once you've collected it, see the chapter on "Cartographic and graphical techniques." For calculating the relationships between your data points, see "Statistical tests."