Fieldwork Methodology

Introduction

Fieldwork is an essential component of A-Level Geography, assessed through a Non-Examined Assessment (NEA) worth 20% of the A-Level. The NEA requires students to design and execute an independent geographical investigation (3,000–4,000 words) that demonstrates the full enquiry process from question formulation through data collection, analysis, and evaluation. This topic covers the key methodologies, techniques, and principles that underpin successful geographical fieldwork.

Key Concepts and Definitions

Term	Definition
Hypothesis	A testable statement predicting the expected outcome of an investigation
Research question	An open question that guides the direction of an investigation
Primary data	Data collected directly by the researcher through fieldwork (e.g., surveys, measurements)
Secondary data	Data collected by someone else for a different purpose (e.g., census data, satellite imagery)
Quantitative data	Numerical data that can be measured and statistically analysed (e.g., river velocity, temperature)
Qualitative data	Non-numerical data describing qualities, perceptions, and experiences (e.g., interview responses, photographs)
Sampling	The process of selecting a subset of a population or area for investigation
Representativeness	The extent to which a sample accurately reflects the characteristics of the whole population or area
Reliability	The consistency of results — would the same method produce the same results if repeated?
Validity	The extent to which the method actually measures what it claims to measure
Bias	Systematic error in data collection or sampling that skews results in a particular direction
Ethical considerations	Principles governing the treatment of human participants, the environment, and data in research
Risk assessment	The identification and management of potential hazards before and during fieldwork
Triangulation	Using multiple methods or data sources to cross-check and strengthen findings
Geographical information system (GIS)	Software that captures, stores, analyses, and displays geographic data
Spearman’s rank correlation	A statistical test measuring the strength and direction of association between two ranked variables
Chi-squared test	A statistical test determining whether there is a significant association between observed and expected frequencies
Mann-Whitney U test	A statistical test comparing the medians of two independent samples to determine if they are significantly different
Standard deviation	A measure of the spread or dispersion of data around the mean

Research Design

Formulating a Research Question or Hypothesis

A good geographical investigation starts with a focused, manageable question or hypothesis.

Characteristics of a good research question:

Geographically grounded (involves spatial patterns, processes, or relationships)
Specific and focused (not too broad)
Testable with available methods and resources
Rooted in geographical theory or concepts
Allows for analysis and evaluation, not just description

Examples:

Hypothesis: “Cross-profile width of the River X increases with distance downstream.”
Research question: “To what extent has regeneration improved the quality of the urban environment in Location Y?”
Hypothesis: “Perceptions of place differ significantly between long-term residents and recent arrivals in Location Z.”

The Geographical Enquiry Process

Identify the question/hypothesis: Based on observation, theory, or secondary data
Plan the methodology: Decide what data to collect, how, where, and when
Collect data: Conduct fieldwork, gathering primary data; supplement with secondary data
Present and analyse data: Use appropriate graphs, maps, and statistical tests
Draw conclusions: Relate findings back to the original question/hypothesis and geographical theory
Evaluate: Assess the strengths, limitations, and reliability of the methodology and findings
Suggest improvements: Propose how the investigation could be refined or extended

Ethical Considerations

Informed consent: Participants must understand what they are being asked to do and agree to take part
Anonymity and confidentiality: Personal data should not be traceable to individuals without their consent
Right to withdraw: Participants must be able to stop at any time
Respect for people and places: Avoid causing damage, disruption, or offence
Data protection: Store and handle personal data responsibly (GDPR compliance in the UK)
Environmental impact: Minimise disturbance to natural environments during fieldwork

Data Collection Methods

Physical Geography Methods

Method	What It Measures	Equipment	Considerations
River velocity	Speed of water flow (m/s)	Flow meter / impeller; float and stopwatch; orange (float method)	Measure at multiple points across the channel; at 0.6 depth for mean velocity
River discharge	Volume of water per unit time (m³/s)	Velocity × cross-sectional area (width × depth)	Requires systematic depth measurements across the channel
Channel cross-profile	Shape and dimensions of the river channel	Tape measure, ranging poles, metre rule	Measure at representative sites; record bed profile
Sediment analysis	Particle size, shape, and roundness	Callipers, Powers’ roundness scale, graduated sieves	Sample systematically; use Wolman sampling (100 particles per site)
Beach profile	Shape and gradient of a beach	Clinometer, ranging poles, tape measure	Measure at regular intervals from the cliff to the low water mark
Cliff profile	Shape and features of a cliff face	Clinometer, tape measure, field sketch	Record bedding planes, joints, vegetation, evidence of mass movement
Sediment sorting analysis	Distribution of particle sizes	Sieve stack, electronic balance	Shake for standard time; weigh each fraction
Microclimate measurements	Temperature, humidity, wind speed, etc.	Thermometer/hygrometer, anemometer, data logger	Record at consistent heights and times; shade from direct sun
Soil infiltration	Rate at which water enters the soil	Infiltrometer (or cut-off pipe and stopwatch)	Record time for water level to drop; compare across different surfaces
Ecological sampling	Species diversity and abundance	Quadrats (0.5m × 0.5m or 1m × 1m), transect lines	Random or systematic placement; identify and count species

Human Geography Methods

Method	What It Measures	Approach	Considerations
Questionnaires	Opinions, behaviours, demographics	Structured questions, face-to-face or self-completed	Pilot the questionnaire; ensure clear, unbiased questions; record sample size and method
Interviews	In-depth perspectives and experiences	Semi-structured or unstructured; recorded (with consent)	Prepare open questions; listen actively; note non-verbal responses
Land use survey	Types and distribution of land use	Systematic recording of land use along transects or in a grid	Use standardised categories; map results
Environmental quality survey	Subjective assessment of environmental quality	Bipolar scales (e.g., -5 to +5) rating specific criteria	Use consistent criteria across all sites; acknowledge subjectivity
Pedestrian count	Footfall and activity levels	Count pedestrians passing a point in a set time	Standard time period; same time of day at each location; consider day of week
Clone town survey	Homogenisation of retail	Record number of independent vs chain stores	Compare across town centres
Index of Multiple Deprivation (IMD) analysis	Relative deprivation levels	Secondary data from government statistics	Understand the domains (income, employment, health, education, etc.)
Photographic evidence	Visual record of change, quality, or features	Systematic photography at designated viewpoints	Note location, direction, date, and time; obtain consent for people photos
Soundscapes	Auditory environment	Record and classify sounds at specific locations	Note sources, intensity, and duration; ethical considerations with recording near private spaces

Sampling Strategies

Strategy	Description	Advantages	Disadvantages
Random sampling	Every item has an equal chance of selection (e.g., using random number generator for grid coordinates)	Eliminates bias; statistically robust	May miss important areas; may cluster by chance
Systematic sampling	Regular, evenly spaced intervals (e.g., every 50 m along a transect, every 10th person)	Simple; ensures coverage; easy to replicate	May coincide with a pattern in the data (e.g., regular housing layout)
Stratified sampling	Divide the population or area into subgroups (strata) and sample proportionally from each	Ensures representation of all subgroups; more accurate	Requires prior knowledge of the population structure
Opportunistic (convenience) sampling	Selecting the nearest or most convenient items	Quick and easy	Highly biased; not representative
Cluster sampling	Randomly select clusters (e.g., streets, grid squares) then sample within them	Practical for large areas	Less statistically rigorous than simple random sampling

Sample size: Larger samples are more representative but require more time and resources. A minimum of approximately 30 data points is generally recommended for statistical validity. For questionnaires, approximately 50–100 responses are in most cases adequate for A-Level investigations.

Statistical Tests

When to Use Each Test

Test	Purpose	Data Type	Conditions
Spearman’s rank (ρ)	Tests for correlation between two variables	Ordinal or continuous; at least 10 pairs of data	Data does not need to be normally distributed
Chi-squared (χ²)	Tests whether observed frequencies differ significantly from expected frequencies	Categorical data (frequency counts)	All expected frequencies ≥ 5
Mann-Whitney U	Tests whether two independent samples are significantly different	Ordinal or continuous; independent samples	Data does not need to be normally distributed
Student’s t-test	Tests whether the means of two samples are significantly different	Continuous, normally distributed data	Requires approximately normal distribution
Mean, median, mode	Measures of central tendency	Any numerical data	Consider outliers affecting the mean
Standard deviation	Measure of spread/dispersion	Any numerical data	Larger SD = more spread; smaller SD = more clustered
Interquartile range	Spread of the middle 50% of data	Any numerical data	Less affected by outliers than standard deviation

Spearman’s Rank Correlation: Step-by-Step

Rank both sets of data separately (assign 1 to the smallest value, 2 to the next, etc.)
Calculate the difference (d) between each pair of ranks
Square each difference (d²)
Apply the formula: ρ = 1 − (6 × Σd²) / (n × (n² − 1))
Compare the calculated value to the critical value at the appropriate significance level (commonly 0.05) 0.05)
If the calculated value exceeds the critical value, reject the null hypothesis — there is a significant correlation

Interpretation: ρ ranges from -1 (perfect negative correlation) to +1 (perfect positive correlation). Values near 0 indicate no correlation.

Chi-Squared Test: Step-by-Step

State the null hypothesis (e.g., “There is no significant association between X and Y”)
Record observed frequencies (O) in a contingency table
Calculate expected frequencies (E) for each cell: E = (row total × column total) / grand total
Calculate χ² = Σ ((O − E)² / E) for each cell
Calculate degrees of freedom: df = (number of rows − 1) × (number of columns − 1)
Compare the calculated χ² to the critical value at the 0.05 significance level
If calculated > critical, reject the null hypothesis

GIS and Data Presentation

GIS Applications in Fieldwork

GIS software (e.g., ArcGIS, QGIS, Google Earth Pro) enables:

Mapping fieldwork data: Plotting sample points, land use, or environmental quality scores on a base map
Spatial analysis: Identifying patterns, clustering, and relationships that may not be apparent in tables or graphs
Overlay analysis: Combining multiple data layers (e.g., flood risk zones and property values) to investigate spatial relationships
Buffer zones: Creating zones around features (e.g., 500 m buffer around a regeneration site) to analyse proximity effects
Digital elevation models (DEMs): Visualising and analysing topography

Effective Data Presentation

Data Type	Appropriate Presentation Methods
Spatial data	Choropleth maps, proportional symbol maps, dot maps, GIS layers, heat maps
Changes over time	Line graphs, bar charts, population pyramids
Comparisons	Bar charts, divided bar charts, scatter graphs
Proportions	Pie charts, proportional circles (use sparingly — often better alternatives exist)
Distributions	Histograms, box-and-whisker plots, frequency polygons
Relationships	Scatter graphs with line of best fit, Spearman’s rank results
Profiles	Cross-section diagrams, beach profiles, cliff profiles
Qualitative data	Annotated photographs, word clouds, quotation extracts, mental maps

Principles of good presentation:

Choose the most appropriate method for the data type
Ensure all graphs and maps have titles, labelled axes, legends, and scales
Use consistent formatting throughout
Annotate key features and trends
Ensure presentation supports analysis, not just description

Risk Assessment

Identifying and Managing Fieldwork Risks

A risk assessment must be completed before any fieldwork. It should identify:

Category	Example Hazards	Mitigation
River/coastal fieldwork	Drowning, slipping on wet rocks, fast currents, tides	Check weather and tide times; never work alone; wear appropriate footwear; establish emergency procedures; stay away from deep water and cliff edges
Urban fieldwork	Traffic, stranger danger, harassment, getting lost	Work in pairs or groups; carry a charged phone; inform someone of your planned route and return time; wear high-visibility clothing near roads
Weather	Hypothermia, heatstroke, sunburn, lightning	Check forecast; wear appropriate clothing; carry waterproofs/sunscreen/water; have a bad-weather contingency plan
Equipment	Cuts from glass thermometers, heavy equipment	Use plastic equipment where possible; carry first aid kit
Data collection	Ethical breaches, conflict with public	Follow ethical guidelines; be polite and respectful; do not photograph people without consent

Risk assessment format: List each hazard, assess its likelihood (low/medium/high) and severity (low/medium/high), then describe the control measures that will reduce the risk to an acceptable level.

Evaluation

Evaluating the Investigation

A strong evaluation considers:

Methodology:

Were the sampling methods appropriate and representative?
Was the sample size adequate?
Were there sources of bias or error?
Were the data collection techniques reliable?

Data quality:

How accurate were the measurements?
Were there anomalies or outliers, and how were they handled?
Was secondary data from a reliable source?

Analysis:

Were the statistical tests appropriate for the data type?
Were the results statistically significant?
Were limitations of the statistical tests acknowledged?

Conclusions:

Do the conclusions follow logically from the evidence?
Are alternative explanations considered?
How do findings relate to geographical theory?

Improvements:

How could the methodology be improved with more time or resources?
What additional data would strengthen the conclusions?
Could different methods provide more reliable or valid results?

Structuring the Evaluation

A good evaluation should:

Identify specific limitations (not generic statements like “the weather was bad”)
Explain how each limitation affected the results
Suggest concrete improvements
Assess the overall reliability and validity of the conclusions
Acknowledge what the investigation did well (balanced evaluation)

Common Pitfalls

Collecting data without a clear question: Data collection should be driven by the research question or hypothesis. Students often collect large amounts of data that they cannot use effectively. Before collecting any data, ask: “How will this help me answer my question?”
Using statistical tests without understanding the conditions: Each test has specific requirements (data type, minimum sample size, distribution). Using a test inappropriately (e.g., chi-squared with expected values below 5) invalidates the results. Always check conditions before applying.
Describing data rather than analysing it: Stating that “the velocity increased from 0.3 m/s to 0.8 m/s” is description. Explaining that “velocity increased downstream due to the smoother channel bed and increased discharge, consistent with the Bradshaw model” is analysis. Always explain patterns and link them to geographical theory.

Worked Examples

Example 1: Planning a River Study

Research question: How does the cross-profile of the River Bollin change with distance downstream?

Methodology plan:

Site selection: Choose 8 sites at approximately equal intervals along the river’s course from source to mouth, ensuring accessibility and safety. Sites should represent upper, middle, and lower course.
Data collection at each site:
- Measure channel width using a tape measure stretched across the channel from bank to bank (wetted perimeter)
- Measure depth at regular intervals across the channel (every 50 cm or 1 m depending on width) using a metre rule
- Record velocity at each depth measurement point using a flow meter at 0.6 depth (for mean velocity)
- Calculate cross-sectional area (width × mean depth) and discharge (area × mean velocity)
- Photograph each site and note bed and bank material
Sampling: Systematic sampling along the river course (sites at equal intervals of distance). At each site, systematic depth measurements at regular intervals across the channel.
Secondary data: Use OS maps to plot site locations and calculate distance from source. Use EA (Environment Agency) data for long-term discharge records.
Presentation: Cross-profile diagrams for each site; scatter graphs of width, depth, and discharge against distance downstream; mapped results on a GIS base map.
Analysis: Spearman’s rank correlation to test the significance of relationships between distance downstream and channel variables. Compare results to the Bradshaw Model predictions.
Risk assessment: Check weather and river conditions; work in groups; wear appropriate footwear; avoid sites with fast current or deep water; carry first aid kit and mobile phone.

Example 2: Planning an Urban Study

Research question: To what extent has regeneration improved environmental quality in the Salford Quays area?

Methodology plan:

Data collection methods:
- Environmental quality survey (EQS) at 10 sites: rate each site on bipolar scales for litter, graffiti, noise, air quality, vegetation, building condition, and perceived safety (-5 to +5)
- Land use survey: record the function of every building or plot along designated transects
- Pedestrian counts: 5-minute counts at each site at consistent times
- Photographic evidence: systematic photographs at each site, before/after comparison using secondary images
- Questionnaires: 30 respondents asked about their perceptions of change, satisfaction, and sense of place
Secondary data: Census data for the area (2011 and 2021), IMD data, property price data (Zoopla/Rightmove), historical maps and photographs from Salford City Council archives.
Sampling: Stratified sampling — select sites to cover regenerated areas (MediaCityUK, The Lowry), partially regenerated areas, and non-regenerated areas for comparison.
Presentation: Choropleth maps of EQS scores, annotated photographs, pie charts of land use, bar charts of pedestrian counts, word clouds of questionnaire responses.
Analysis: Mann-Whitney U test to compare EQS scores between regenerated and non-regenerated areas. Thematic analysis of questionnaire responses. Comparison with secondary data trends.
Evaluation: Acknowledge subjectivity of EQS, limited sample size for questionnaires, difficulty isolating regeneration effects from other factors.

Summary

Fieldwork follows a structured enquiry process: question → plan → collect → present → analyse → conclude → evaluate.
Both primary and secondary data, and both quantitative and qualitative methods, strengthen an investigation through triangulation.
Sampling strategy must be chosen carefully to ensure representativeness and minimise bias.
Statistical tests (Spearman’s rank, chi-squared, Mann-Whitney U) enable rigorous analysis of data patterns and relationships.
GIS enables spatial analysis and effective data presentation.
Risk assessment and ethical considerations must be addressed before fieldwork begins.
Evaluation should identify specific limitations, explain their impact, and suggest concrete improvements.
The strongest investigations link findings explicitly to geographical theory and concepts.

Sources: AQA Geography (7037) specification; AQA NEA guidance documents; Clark, Skills in Geography (2017); Harris and Jenner, Geographical Skills and Fieldwork (2016); Environment Agency data services; OS mapping guidelines.

Mark this page as reviewed