Planning fieldwork and the fieldwork question

What the fieldwork report is

The internal assessment in geography is a written report of up to 2,500 words based on fieldwork: an investigation of one focused question about a real local place, answered with information collected first-hand. It is the same at SL and HL, it is marked out of 25 by your teacher against six criteria (A to F), and a sample of reports from each school is moderated by the IB. About 20 hours of teaching time are set aside for it, from planning to the final report.

Three rules shape everything else. First, the investigation must be based on primary data, collected in the field by measuring, counting, observing or asking; secondary data such as census figures or gauge records may support it but only in a small way. Secondly, the question must be linked to a topic in the course, and the report says which. Thirdly, data collection can be shared, but the writing is yours alone.

Secondary sources still have a place: a gauge record can show whether the river was high or low on the day of fieldwork, and a census table can show whether a surveyed street is typical of its town.

The criteria and the shape of the report

The report is marked against six criteria, and the most efficient structure follows them in order, one section each. The suggested word counts are guidance, not rules, but they show where the marks are: the written analysis alone is worth almost a third of the total.

Choosing a topic

A good topic starts from something in the course that can be seen and measured locally. The question then applies a model, a theory or a process to a real place: how a river channel changes downstream, how a groyne affects a beach, how environmental quality changes away from a city centre, how tourists perceive a crowded beauty spot. The topic can be physical, human, environmental or a mix of these.

Some topics cannot be studied at the local scale with primary data and are unlikely to be suitable: ocean–atmosphere interactions, the characteristics of extreme environments, geophysical systems, tourism and sport at the international scale, measuring food and health, and the core units on population and global climate. Most of the HL extension works at the global scale too; local responses to global interactions and local and global resilience are the exceptions. A global issue works only if it is reduced to something local, such as resource stewardship measured in one town.

Writing the fieldwork question

The fieldwork question is the single question the whole report answers. It decides what data to collect, which methods are relevant and what the conclusion must say. A strong question is focused, geographical (it is about variation over space, or a process in a place), names a precise location and can be answered with primary data in the time available.

Questions that compare places or measure change along a line work particularly well, because they give a spatial pattern to analyse and enough sites for a statistical test. "To what extent…" questions also suit the criteria, since they invite a nuanced conclusion. A question with a yes/no answer ("Is the river wider downstream?") is usually too narrow: rephrase it to ask how, or how far, something varies.

Hypotheses and predictions

A hypothesis turns the question into testable statements. Each one predicts a relationship or difference between two variables and comes from geographical theory, which is why criterion A asks for the link to theory: the theory is what allows a prediction to be made. A hypothesis should name both variables and the direction of the expected change ("increases", "decreases", "is higher at").

Statistical tests work with a null hypothesis, written H₀, which states that there is no relationship or no difference. The test calculates how likely the observed result would be if H₀ were true; if that is very unlikely, H₀ is rejected and the alternative hypothesis (H₁, your prediction) is accepted at a stated level of confidence.

The worked investigation: the Kerran Beck

One investigation is followed through every stage, from question to evaluation, using hypothetical data for an invented upland stream, the Kerran Beck. Its question is: How do channel characteristics change with distance downstream along the upper 10 km of the Kerran Beck? The theory behind it is the Bradshaw model, which predicts how a river’s channel, flow and load change from source to mouth.

H3 is the most interesting of the four, because it contradicts what many people expect: water looks fast in a steep mountain stream. The prediction rests on hydraulic radius, the cross-sectional area divided by the wetted perimeter, which measures how efficiently a channel carries water. Testing a prediction that goes against common sense gives the analysis more to explain.

The geographic context and the location map

The context section explains, briefly, where the investigation takes place and why the place suits the question. About 300 words are enough for the question, the context, the theory and the hypotheses together. It locates the site (region, river or settlement, with distances and directions), describes the features that matter (geology, land use, a weir or a tributary, a groyne field, a shopping street), and gives the link to the course topic and to the theory.

The location map is required for the top band of criterion A, and a missing or poor map is a common reason for losing a mark there. It should be your own map, drawn by hand or with GIS, showing the study area at a scale where every site can be seen and numbered, with an inset locating the area within its region or country. A downloaded map with hundreds of irrelevant details, no sites and no conventions earns little.

Planning the data collection

Before going into the field, list every variable each hypothesis needs, how it will be measured, with what equipment and at how many sites. This plan becomes the methods section (criterion B), where each method is described and justified: why it was chosen and how it helps answer the question. Data are quantitative data (numbers from measurements and counts) or qualitative data (words, images and opinions), and many good investigations combine the two.

Qualitative data have their own value. Interviews explain why people behave as counts show they do, and an annotated photograph records the evidence for a judgment, such as the condition of a building. They can be coded into categories and counted when they need to be compared.

Time, sites and circumstances

Record when and in what conditions the data were collected: the date, the time, the weather of the previous days and the river level. A river measured two days after heavy rain gives quite different values from the same river in a dry spell, and all sites should be measured within as short a period as possible so that the results can be compared. In human geography the same applies to the day of the week, the time of day and events such as markets or school holidays.

The pilot study

A pilot study is a short trial of the methods before the main collection, at one or two sites or with a few respondents. It shows whether the equipment works, how long each site takes, whether the sites are safe and accessible, and whether questionnaire questions are understood as intended. Changes made after the pilot, and the reasons for them, are worth a sentence in the methods section and in the evaluation.

Risk assessment and ethics

Fieldwork takes place in real environments, from riverbeds to busy streets, and every investigation needs a risk assessment before it starts. Schools follow their own risk-assessment procedures, and the teacher approves the plan; the student’s part is to identify the hazards of the chosen sites and methods and to plan how to reduce them.

Ethics are part of the same planning. People who answer a questionnaire or an interview give informed consent: they are told who you are, what the survey is for and that they may refuse or stop at any time. Answers are kept anonymous, and no one is photographed in a way that identifies them without permission. Private land needs the owner’s permission, and fieldwork should leave no damage: pebbles returned to the river, vegetation not trampled, wildlife not disturbed.

Three more investigations from question to plan

The same planning steps apply to every kind of topic. Three invented outlines, one coastal, one urban and one on tourism, show how a question, its theory, its hypotheses and its methods fit together. All are invented examples, and each would need its own location map and risk assessment.

A coastal investigation: groynes and longshore drift

Question: How does the groyne field at the north end of a named beach affect beach width and sediment size along the beach? Theory: longshore drift moves sediment along the coast in the direction of the dominant waves, and a groyne traps sediment on its updrift side, starving the downdrift side. Hypotheses: beach width is greater on the updrift side of each groyne than on the downdrift side; mean sediment size is larger on the downdrift side, where finer material has been removed.

Methods: beach profiles on both sides of five groynes, measured at the same state of the tide; sediment samples of 20 pebbles at fixed points on each profile; wave direction recorded over the day; a field sketch of the groyne field. The data allow a paired comparison of updrift and downdrift sides and a test of difference between them.

An urban investigation: environmental quality from the centre

Question: How do environmental quality and land use change along a 3 km transect from the CBD of a named town to its southern suburbs? Theory: urban land-use models and the bid-rent idea predict commercial land at the centre and residential land outwards, with inner-city areas of older, denser housing that may score lower on environmental quality. Hypotheses: the share of commercial land use decreases with distance from the CBD; environmental quality scores increase with distance from the CBD.

Methods: systematic sites every 250 m along the transect; land use of every building frontage within 50 m of each site; an environmental quality survey on eight criteria at each site, scored by two people independently to reduce bias; a pedestrian count of ten minutes at each site; annotated photographs. The results suit a compound bar graph, a bipolar graph and a correlation test against distance.

A tourism investigation: pressure at a honeypot site

Question: How do visitor numbers, path erosion and perceptions of crowding change with distance from the car park along the main path at a named honeypot site? Theory: visitor pressure decreases with distance from access points, and carrying capacity is exceeded first where visitors concentrate. Hypotheses: path width and depth of erosion decrease with distance from the car park; the percentage of visitors who feel the site is crowded decreases with distance.

Methods: visitor counts at five points along the path at the same times on a weekday and a weekend day; path width and erosion depth measured every 100 m; a short questionnaire of visitors at three points, with a Likert scale for crowding. Mixing counts, measurements and perceptions gives an integrated investigation that links physical change to human behaviour.

Using secondary sources and references

Secondary sources play a supporting role. They can set the context (the population of the town, the geology of the drainage basin), allow a comparison (river gauge data for the day of fieldwork against the long-term average), or confirm the base map. They should never replace the primary data or make up a large part of the analysis.

Every source used is referenced in the text and listed at the end in a standard system such as Harvard: the author or organisation, the year, the title, the publisher or website, and for online sources the URL and the date accessed. Images and maps are credited in their captions, including base maps adapted from an online service. References, the title page, the contents page, map keys and labels of ten words or fewer are not counted in the word limit.

A timetable for the 20 hours

The first draft is the only one your teacher can comment on, so it should be complete: every section written, every figure in place. A partial draft wastes the one chance for advice.

Planning checklist

• To do:

  My question is focused, geographical and names a precise local place.

• To do:

  The question can be answered with primary data that I can collect safely in the time available.

• To do:

  I can name the course topic and the theory or model the question tests.

• To do:

  I have three or four hypotheses, each with two measurable variables and a direction, and matching null hypotheses.

• To do:

  I have planned the variables, methods, equipment, sites and sampling, and can justify each.

• To do:

  I have carried out a pilot study and adjusted the methods where needed.

• To do:

  A risk assessment has been completed and approved, and I know how I will obtain informed consent.

• To do:

  I have drafted a location map with a title, north arrow, scale, numbered sites, key and inset.

• To do:

  I have recorded the date, time and conditions of data collection.