Every IB Geography student writes one fieldwork report based on primary data you collect yourself. It is worth a big slice of your grade, so getting the method and analysis right really counts.
This lesson walks the whole pipeline and drills the numbers you may need — sampling interval, Spearman's rank, and percentages. Press Start.
IA · overview
What the report must do
The IA is a written fieldwork report of around 2,500 words, built on a fieldwork question you investigate using primary data linked to the syllabus.
It is marked against IB criteria covering the question & geographic context, method, quality & treatment of information, written analysis, and conclusion & evaluation.
The report should read as a genuine enquiry: a clear aim, evidence, and an honest judgement of how well you answered it.
Golden rule: the whole report exists to answer one focused question. Every graph, map and stat must earn its place by helping to answer it.
Stage 1 · the question
A good fieldwork question
The strongest questions are focused, measurable, geographic and feasible in the time and place available. They usually contain a clear variable to test.
Link it to a concept or model (e.g. the Bradshaw model for rivers, a sphere of influence for shops, an urban land-use model).
Phrase it so it can be answered with data, not just described.
Example shapes: "How does river velocity change downstream?" or "Does pedestrian footfall decline with distance from the CBD?" — each names a variable that can be measured and tested.
Quick check
Pick the best question
?Which of these is the strongest IB fieldwork question?
Stage 2 · collecting data
Primary & secondary data
Primary data is what you gather yourself in the field (measurements, counts, questionnaires, environmental quality surveys). Secondary data comes from published sources (census, maps, official statistics) and gives context.
Before the field: a risk assessment and a pilot survey (a small trial run) let you fix problems with your method before you collect the real data.
Stage 2 · sampling
Sampling strategies
You cannot measure everything, so you sample. The main strategies are:
Random — every item has an equal chance (removes bias, but may cluster).
Systematic — take every nth item, or points at fixed intervals along a transect.
Stratified — sample sub-groups in proportion, so each is represented.
sampling interval = population size ÷ sample sizefor systematic sampling: how many items to skip between each one you pick
Calculate
Your turn — sampling interval
1A list has 200 possible survey sites. You want a systematic sample of 20. Calculate the sampling interval (pick every nth site).
nth
Hint: interval = population ÷ sample = 200 ÷ 20.
Quick check
Name the strategy
?You measure the river channel at points every 500 m along its length. Which sampling strategy is this?
Stage 3 · presenting data
Choosing presentation techniques
Marks come from choosing appropriate, varied techniques and using them well:
Located techniques on a base map — proportional symbols, flow lines, choropleth shading.
Graphs — scatter graphs to show relationships; bar and line graphs for patterns.
Every figure needs a title, key, units and a sentence saying what it shows.
Avoid: a wall of identical bar charts. Match the technique to the data type and to the question.
Sort it
Which stage of the report?
Tap a task, then tap the stage of the report it belongs to.
📋 Method
📊 Present & analyse
🧭 Evaluation
Stage 4 · statistical analysis
Spearman's rank correlation
To test whether two variables are related, geographers use Spearman's rank correlation coefficient (rs):
rs = 1 − ( 6 Σd² ) ÷ ( n(n² − 1) )d = difference between the two ranks of each pair · n = number of pairs
The result runs from +1 (perfect positive) through 0 (no correlation) to −1 (perfect negative). You then check it against significance tables.
Worked example
For n = 5 pairs, suppose the squared rank differences add up to Σd² = 4.
The sign tells you the direction, the size the strength:
Close to +1: strong positive — as one variable rises, so does the other.
Close to −1: strong negative — as one rises, the other falls.
Near 0: little or no correlation.
Critical point: correlation is not causation. A significance test tells you how likely the pattern is due to chance, but you must still explain the geography behind it.
Quick check
What does r_s = +0.8 mean?
?Your analysis gives rs = +0.8. What does this best indicate?
Stage 4 · normalising data
Percentages & fair comparison
Raw counts can mislead if sample sizes differ, so we often convert to percentages to compare fairly:
percentage = (part ÷ total) × 100e.g. how many of those surveyed chose a particular response
Tip: stating a percentage and the sample size (n) is best practice — "65% of 40 respondents" is far stronger than "most people".
Calculate
Your turn — percentage
3You survey 40 people; 26 say the regenerated square feels safer. What percentage is that?
%
Hint: (26 ÷ 40) × 100.
Match it
Match the term to its meaning
Tap a description on the left, then its matching term on the right.
Description
Term
Stage 5 · conclusion & evaluation
Conclusion & evaluation
The report ends by answering the question with evidence, then honestly evaluating the enquiry:
State a clear conclusion that refers back to the original question and to your data.