A good graph choice matches the type of data and the question in the aim. Two measured variables and a question about a relationship point to a scatter graph. Parts of a whole point to a pie chart or a divided bar, and compass directions point to a wind rose.
Questions on this appear whenever you must present or evaluate data from fieldwork.
This lesson follows processing a small dataset in fieldwork analysis and evaluation.
How do you match a graph to the data?
Ask two questions. First, what kind of data is it: counts of categories, measurements, proportions or directions? Second, what does the aim want to know: a comparison, a relationship, a share or a pattern around a circle?
| What the aim asks | Data you have | Suitable graph |
|---|---|---|
| Does one variable change with another? | Two measurements per site | Scatter graph with line of best fit |
| Which category is largest? | Counts for named categories | Bar chart |
| How are sizes spread? | Measurements grouped into classes | Histogram |
| What share does each part have? | Percentages of a whole | Pie chart or divided bar |
| Which directions are most frequent? | Compass direction counts | Wind rose |
Worked example
The invented aim is: “Pedestrian counts decrease with distance from the town centre.” The data is the seven sites from the previous lesson.
Step 1, data type: distance (km) and pedestrian count are both measurements, one pair per site.
Step 2, aim type: the aim asks about a relationship.
Step 3, choice: a scatter graph, with distance on the horizontal axis and count on the vertical axis.
Step 4, scales: distance runs from 0.1 to 2.0, so use 0 to 2.0 km with 1 cm for each 0.2 km. Counts run from 35 to 84, so use 0 to 90 with 1 cm for each 10.
Step 5, plot and fit: plot each site as a cross, then draw a line of best fit sloping down from the upper left to the lower right. Label both axes with units and give the graph a title that repeats the aim.
The points fall close to the line, which supports the aim. The conclusion itself is covered in linking a conclusion to the investigation aim.
What mistake costs marks?
A common error is to pick a graph by habit rather than by aim.
Mistaken choice: the student draws a line graph joining the seven sites in order and calls it a time series, because the sites are labelled A to G.
The sites are not points in time, and letters are only labels. The data is two measurements, so a scatter graph with a line of best fit is the right display. A bar chart would hide the distances, because bars suggest equal gaps.
Check yourself
All data is invented.
1. A class wants to show which of five land uses is most common in a street survey. Name a suitable graph.
Show answer
A bar chart. The land uses are named categories with a count for each, and the question is about which is largest.
2. A student records the wind direction each day for 30 days. Which graph shows how often each direction occurred?
Show answer
A wind rose. It uses the compass circle, so neighbouring directions stay beside each other.
3. Land use in a district is 40% housing, 30% farmland and 30% other. What angles would you use on a pie chart?
Show answer
Multiply each percentage by 3.6: 40 × 3.6 = 144°, 30 × 3.6 = 108° and 108°. Check: 144 + 108 + 108 = 360°.
Where this leads next
Once a graph is drawn, you will see points that do not fit. The next lesson shows how to identify anomalous evidence transparently. The fieldwork sample and graph planner helps you rehearse graph choice and scales on practice data.
If you often pick the wrong graph for the aim, a teacher in online one-to-one Geography tuition can ask why you chose each one and help you fix it.