Have you ever tried to remember exactly what you saw at a beach five hours after leaving? Good fieldwork starts with a well-designed data sheet. Every sheet must include essential metadata at the top: the site name, GPS coordinates, date, time, and weather conditions.
Sheets should also include brief instructions so everyone records data consistently, and space for annotated sketches to capture extra context. A well-designed sheet ensures validity, meaning your method actually measures what you set out to investigate.
You cannot possibly measure every single pebble on a beach, so geographers must choose a small selection to study. A sampling strategy is the method used to select a representative subset of data. OCR focuses on three main approaches to avoid bias:
Working out exactly where to stop along a transect requires a simple calculation. To find your systematic sampling interval, divide the total length of your transect by the number of samples you need to collect.
A student needs to take 8 sediment samples along a 120m stretch of coastline. Calculate the interval between each sample.
Step 1: Identify the total length and the number of samples required.
Step 2: Substitute the values into the formula.
Step 3: Calculate the final answer with units.
Why does asking just two people in a town centre give a weak conclusion? Taking a larger sample size makes your data a representative sample, meaning it accurately reflects the whole area.
Hitting the bullseye on a dartboard is a great shot, but true skill is hitting it three times in a row. In fieldwork, accuracy is how close a measurement is to the true value, free from human or equipment error. You can improve accuracy by upgrading equipment, such as swapping a floating orange peel for a digital flow meter, or using calipers instead of a ruler to measure pebbles.
Reliability is about reproducibility and consistency. Data is reliable if someone else could repeat your exact method and get the same results. You can improve reliability by taking repeat measurements at the same site and calculating a mean to smooth out temporary fluctuations.
When collecting subjective data like an EQS, you can boost inter-observer reliability by working in pairs to agree on a consensus score, or by taking the mode of several students' opinions. Always follow standardised procedures, such as measuring river velocity at exactly 0.6 of the total depth every time.
To prove a new sea wall is stopping erosion, you need to know what the erosion looks like without one. In geography, a control group (or comparison site) is a baseline location where the independent variable you are studying is missing.
Using a control group is essential because it proves that any changes you observe are actually caused by your specific variable, rather than just natural variation over time.
Students frequently confuse accuracy with reliability. Remember: better equipment improves accuracy, while repeating measurements improves reliability.
When evaluating an unfamiliar data collection sheet in Paper 3, look for missing metadata (like units or location coordinates) and the lack of a 'total' column.
If an exam question asks you to justify your sample size, you must use the term 'representative' to secure the full marks.
Suggesting 'use a digital flow meter instead of a float' is a classic OCR mark scheme point to suggest an improvement in data collection accuracy.
Tally charts
A method of recording frequency data using marks, grouped in fives, to quickly count distinct categories.
Bipolar scale
A qualitative measuring scale featuring opposing descriptive words at either end (e.g., -3 to +3), commonly used in Environmental Quality Surveys.
RICEPOTS
A classification system used in land-use mapping representing Residential, Industrial, Commercial, Entertainment, Public, Open space, Transport, and Services.
Validity
The extent to which a data collection method actually measures what it was intended to measure.
Sampling strategy
The systematic, random, or stratified method used to select a representative subset of data from a larger population.
Bias
A distortion in data collection that results in an unrepresentative set of results, such as only surveying one specific age group.
Random sampling
A sampling method where every member of the population or point in an area has an equal chance of being selected.
Systematic sampling
A sampling method where data is collected at regular, pre-defined intervals, such as every 10 metres along a transect.
Stratified sampling
A sampling method where the study area is divided into sub-groups based on known characteristics, and sampled in proportion to their size.
Representative sample
A subset of data that accurately reflects the characteristics of the whole population or area.
Anomalies
Unexpected outlier results that do not fit the general trend or pattern of the data.
Accuracy
The degree to which a measurement reflects actual reality, often improved by using more precise equipment.
Reliability
The extent to which a data collection method produces consistent, reproducible results when repeated.
Inter-observer reliability
The degree of agreement between different people recording subjective data, improved by working in pairs to reach a consensus score.
Control group
A baseline site used for comparison where the specific variable being studied is absent, helping to prove what causes observed changes.
Independent variable
The factor you change or choose to compare in an investigation (e.g. distance from the river source).
Mean
The average value calculated by adding multiple readings and dividing by the number of readings to smooth out fluctuations.
Mode
The most frequently occurring value in a data set, used to find a consensus among several students' scores.
Put your knowledge into practice — try past paper questions for Geography B
Tally charts
A method of recording frequency data using marks, grouped in fives, to quickly count distinct categories.
Bipolar scale
A qualitative measuring scale featuring opposing descriptive words at either end (e.g., -3 to +3), commonly used in Environmental Quality Surveys.
RICEPOTS
A classification system used in land-use mapping representing Residential, Industrial, Commercial, Entertainment, Public, Open space, Transport, and Services.
Validity
The extent to which a data collection method actually measures what it was intended to measure.
Sampling strategy
The systematic, random, or stratified method used to select a representative subset of data from a larger population.
Bias
A distortion in data collection that results in an unrepresentative set of results, such as only surveying one specific age group.
Random sampling
A sampling method where every member of the population or point in an area has an equal chance of being selected.
Systematic sampling
A sampling method where data is collected at regular, pre-defined intervals, such as every 10 metres along a transect.
Stratified sampling
A sampling method where the study area is divided into sub-groups based on known characteristics, and sampled in proportion to their size.
Representative sample
A subset of data that accurately reflects the characteristics of the whole population or area.
Anomalies
Unexpected outlier results that do not fit the general trend or pattern of the data.
Accuracy
The degree to which a measurement reflects actual reality, often improved by using more precise equipment.
Reliability
The extent to which a data collection method produces consistent, reproducible results when repeated.
Inter-observer reliability
The degree of agreement between different people recording subjective data, improved by working in pairs to reach a consensus score.
Control group
A baseline site used for comparison where the specific variable being studied is absent, helping to prove what causes observed changes.
Independent variable
The factor you change or choose to compare in an investigation (e.g. distance from the river source).
Mean
The average value calculated by adding multiple readings and dividing by the number of readings to smooth out fluctuations.
Mode
The most frequently occurring value in a data set, used to find a consensus among several students' scores.