Presenting & Analysing Fieldwork
This lesson covers the general fieldwork skills, terminology, and exam-paper expectations that apply to every student. The specific sites, data, and methods your class actually used will be different — for the case study you’ll write about in the exam, refer to your teacher’s notes and the data you collected on the day.
From Numbers to Meaning
Once your data sheets come back from the field, the next stages of the enquiry cycle are processing and presenting the evidence and then analysing it. The goal is to turn a column of raw numbers into a clear pattern that the reader can see, then explain what the pattern means and how strong it really is.
Two skills matter here. First, you must pick the right kind of graph or map for the data you have. Second, you must use the right kind of statistic to summarise it. The wrong choices hide the pattern; the right ones reveal it.
Match the graph to the data type. Continuous data over time = line graph. Categories = bar chart. Bivariate (two variables) = scatter graph. Spatial pattern = choropleth or proportional symbol map. Frequency distribution = histogram.
Choosing the Right Graph
A bar chart compares totals between separate categories — for example, the number of shops in each retail zone of a high street. The bars do not touch because the categories are independent. A line graph joins data points with a line because the variable changes continuously over time, such as a hydrograph showing river discharge during a storm.
A scatter graph shows the relationship between two variables — for example, distance downstream against river width. A clear upward or downward trend suggests a positive or negative correlation. A histogram shows how often values fall into equal class intervals, such as how many pebbles fall into each 10 mm size band. The bars touch because the data is continuous.
A choropleth map shades areas by value — darker shades for higher values such as deprivation index by ward. They are good for showing patterns across regions but hide variation within a single shaded area. A proportional symbol map uses circles or squares scaled to the value, useful for migration totals or city populations. A flow-line map uses arrows whose width matches the size of the flow, useful for traffic counts or river discharge between gauging stations.
For a high-street pedestrian-flow study, you might place proportional circles on a sketch map of the town — one circle per count point, scaled to the number of pedestrians per minute. The size differences between circles make the pattern obvious at a glance.
Statistics: Central Tendency and Spread
Three measures summarise the ‘middle’ of a data set. The mean is the average. The median is the middle value once the data is sorted. The mode is the most frequent value or category. Each suits different situations: the mean is sensitive to extreme values (one boulder pulls a pebble-size mean upwards), so the median is often a fairer measure for skewed natural data.
You need to know mean, median, mode, range, quartiles and inter-quartile range. The inter-quartile range (IQR) measures the spread of the middle 50% of values, ignoring extremes.
Spread tells you how varied the data is. The range is largest minus smallest, but a single odd reading can stretch it. The inter-quartile range (IQR) is the difference between the upper and lower quartile — the middle 50% — so it ignores outliers and gives a fairer picture. A small IQR means readings cluster tightly around the median; a large IQR means they vary widely.
Identifying Patterns and Trends
Analysis means describing what the graph or map shows and asking why. Useful sentence frames are: ‘The data shows a clear positive correlation between X and Y.’ ‘The pattern is strongest at site 4 and weakens towards site 7.’ ‘Most readings cluster between 25 and 40 mm, with one anomaly at 95 mm.’ Always include numerical evidence from your own results — not just ‘there is a trend’.
An anomaly is a value that does not fit the trend. Geographers do not ignore anomalies — they investigate them. A river-width reading much wider than expected might mean a meander or human modification. A pebble much larger than its neighbours might have been dropped from a cliff fall rather than transported by longshore drift.
Strong analysis links the pattern back to the geographical theory you used to write the hypothesis. If pebble size decreases downstream, that fits the Bradshaw model: attrition smooths and breaks pebbles as they are transported, so they get smaller and rounder away from the source. If pedestrian flow drops with distance from the town centre, that fits the bid-rent model: peripheral shops attract fewer visitors than the central honeypot.
Where the data does not fit the theory, say so. The specification stresses that ‘geography can be messy’ — conclusions that admit unexpected results, and try to explain why, score higher than ones that pretend the data was perfect.
Drawing Lines and Trends
On scatter graphs, sketch a line of best fit by eye so the points are roughly balanced above and below it. The slope shows the direction and the rate of change. How tightly the points cluster around the line shows the strength of the relationship. A line of best fit also lets you interpolate — reading off a value within the data — or extrapolate beyond the data. Extrapolation is risky because the trend may not continue.
Quick Self-Check
Before you move on to conclusions, ask three questions. Did you choose the right type of graph for the data? Did you use the right summary statistic, and explain why? Did you describe the pattern with figures from your own results, including any anomalies? If yes to all three, your analysis is on track.