How to write the IB Geography IA

The Geography IA is a fieldwork report built on primary data you collect yourself, and it is worth a quarter of your final Geography grade at Standard Level and a fifth at Higher. Most students lose marks here long before they write a word, by choosing a question that cannot be answered with their own data or that has no clear location. The rest lose them at the other end, by describing their results instead of analysing them. This guide walks you through the whole thing: what the report is, how it is marked across its unequal strands, exactly how to write each part, and what separates a top-band report from an average one.

The thing to understand first is that the Geography IA is not a research essay with some data attached; it is an account of an investigation you carried out, from question to evaluation, resting at every point on measurements you made in the field. That single fact shapes every decision. It means the question has to be small enough to answer with the data a student can realistically gather in a day or two. It means the location has to be real, accessible and named. And it means that, by the time you reach the analysis, every claim you make should be traceable back to a number you wrote down yourself at a particular site. Students who treat the IA as a chance to write broadly about a geographic issue almost always drift away from their own data and lose marks across several strands at once.

It also pays to see, from the outset, how lopsided the mark scheme is. The five strands are not worth the same, and the written analysis alone is worth roughly as much as two of the other strands combined. This has a practical consequence: the earlier strands — the question, the methods, the presentation — exist largely to make a strong analysis possible. A beautiful set of maps that is never interpreted, or a clever sampling design that the analysis never draws on, leaves most of the available marks untouched. Plan the whole report backwards from the analysis: decide what you will need to be able to say about your data, and then make sure the question, the sampling and the presentation are built to let you say it.

The IB Geography IA at a glance

The fieldwork report is a single investigation of roughly 2,500 words, identical in task and length whether you sit Geography at Standard or Higher Level. It is marked out of 25, and the marks are not spread evenly: the written analysis carries far more weight than any other strand. Because the task is the same at both levels, only the weighting shifts — the IA accounts for 25% of the final grade at SL and 20% at HL. The defining requirement is that the whole report rests on primary data you gathered in the field at a real, locatable site, not on figures downloaded from a database.

The word count is tight for a report that has to carry maps, graphs and statistics as well as prose, so be ruthless about where the words go. Figures, tables and labelled diagrams do a great deal of work that you do not then need to repeat in sentences; a well-designed graph can replace a paragraph of description and free those words for analysis, which is where the marks are. Keep the methods section efficient — enough for another student to repeat your fieldwork, no more — and resist the temptation to narrate the day out. The examiner does not need to know that it rained at lunchtime unless the rain affected your readings, in which case it belongs in the evaluation, not the method. Spend the words you save on interpreting what the data means.

How the Geography IA is marked: the five strands

Marks come from five strands, and they are deliberately unequal. Treat the written analysis as the centre of gravity — it is worth roughly /10, more than any other strand — and make sure the earlier strands feed it cleanly. Build the report strand by strand.

How to write a Geography IA, step by step

1. Pose a focused, locatable question tied to a model. It must be answerable with primary data, anchored to a specific site, and linked to a geographic concept, model or theory that frames what you expect to find.
2. Justify a sampling strategy. Decide how, where and how much to sample so the data is representative, and explain in writing why that strategy suits your question and location.
3. Collect primary data — with a risk assessment and ethics. Carry out the fieldwork yourself, document a risk assessment, and address ethical issues such as consent and environmental impact.
4. Process with maps, graphs and statistics. Turn raw readings into appropriate maps, graphs and statistical measures chosen to fit the data and reveal the pattern.
5. Write a detailed analysis linked to the theory. Interpret the processed data against the question and the model, explaining what the patterns mean rather than restating them — this is where most of the marks sit.
6. Conclude. Answer the fieldwork question directly from the evidence you have presented.
7. Evaluate the method's reliability. Weigh the limitations of your sampling, instruments and conditions, and propose realistic improvements.

Choosing the question is the decision that makes or breaks a Geography IA, so it is worth slowing down over. The ideal question has three properties at once: it is tied to a recognised geographic model or theory, so the analysis has something to test the data against; it is answerable with primary data a student can collect, so you are not reliant on figures you cannot gather; and it is locatable, so the whole investigation has a place. Questions about the structure of a town centre, the change in a river downstream, microclimate variation across an open space, or the decay of an environmental quality index with distance all tend to work, because each pairs a measurable variable with an established model. If you cannot name the model your question will be read against, or the variable you will measure, the question is not yet ready.

The methods strand is really an argument that your data can be trusted. The key word is sampling: you almost never measure everything, so you choose a sample, and you must justify that the sample is representative and free of obvious bias. Explain why you sampled where you did, how many sites or readings you took, and how you decided spacing or timing — a systematic transect, a stratified sample across zones, a random selection within a frame. Wrap in a genuine risk assessment and a note on ethics, including consent for any questionnaires and care for the environment you are studying. None of this needs to be long, but skimping here weakens every claim the analysis later tries to make.

Presentation is the bridge from raw fieldwork to evidence, and it is where students most often leave easy marks on the table. The instinct is to paste in the data table and move on; the better move is to process the data into the form that best reveals the pattern — a located map for anything spatial, a graph for a trend, a statistical measure such as a correlation coefficient to quantify a relationship. Choose techniques that suit the data rather than reaching for the most elaborate one available, and make every figure legible: titled, scaled, labelled and located. A good presentation section sets up the analysis so well that the interpretation almost writes itself.

The written analysis is the heart of the report and the strand worth the most, so give it the most words and the most care. Its job is to explain, not describe. For every pattern in your processed data, say what it is, then say why — and tie that "why" back to the model your question invoked. The strongest analyses make a point of their anomalies: the site that does not fit the trend is often the most revealing part of the whole investigation, because explaining it forces you to reason about local factors the model leaves out. Throughout, keep referring to your own figures, maps and statistics, so the reader can see that the interpretation is grounded in the evidence you collected rather than in general knowledge.

The conclusion and evaluation close the loop. The conclusion should answer the original question plainly, in light of the evidence, and resist the urge to overstate what one day's fieldwork can show. The evaluation then turns a critical eye on the method itself: how reliable was the sampling, how accurate the instruments, how representative the conditions on the day, and what would you change to do it better. A precise evaluation that names a real weakness in your own design — and explains how it could have skewed the result — is far more valuable than a list of generic improvements, and it is what separates a top-band report from a competent one.

Geography IA structure: what goes in each section

The fieldwork report follows the order of the marking strands, which doubles as a clear structure:

• Fieldwork question & geographic context — the locatable question, the model or theory behind it, and a justified methodology.
• Methods of information collection — the sampling strategy, the primary data collected, the risk assessment and ethics.
• Treatment & presentation of information — the processed data shown as maps, graphs and statistics.
• Written analysis — a detailed interpretation tying the results to the question and the theory (the largest section).
• Conclusion & evaluation — the answer to the question and a critical evaluation of the method's reliability.

Keep the primary data visible throughout: every figure in the analysis should be traceable to a map, graph or table earlier in the report, and every presentation technique should earn its place by feeding the analysis that follows.

What a strong vs weak Geography IA looks like

The fastest way to lift your marks is to see the difference. The pairs below take the same fieldwork and show the weak and strong version of each move, so you can see exactly where the marks turn. Read them as patterns to copy rather than content to borrow: the skill on display — locating, processing, analysing against theory — transfers to any fieldwork you choose.

The fieldwork question

Presentation

The written analysis

Common mistakes that cost marks

• A question with no location. If the enquiry is not pinned to a real, namable site, it cannot be a fieldwork report — locate it.
• A question primary data can't answer. If you would have to download the data, the question is wrong for this IA.
• Too little data or biased sampling. A flawed sampling strategy caps the methods strand and weakens everything built on it.
• Raw data left unprocessed. Tables with no maps, graphs or statistics throw away the treatment-and-presentation marks.
• Describing instead of analysing. The dominant strand rewards interpretation linked to theory, not a restatement of the numbers.
• No evaluation of the method. A conclusion that never questions the reliability of the data cannot reach the top band.
• Going over the word count. Material beyond roughly 2,500 words is not rewarded — keep the report tight.

Geography IA — frequently asked questions

Guidance written by experienced IB examiners and aligned to the current Geography guide. Not affiliated with or endorsed by the International Baccalaureate Organization.