Choose the right chart for the data type and question.
Tables
A structured grid of values. Use to present raw or summary data clearly. Include row and column headings, units (e.g. seconds, marks), and a caption that explains what the table shows. Round numbers to a sensible precision (e.g. 2 d.p. for means).
Bar charts
Use for categorical (nominal) data — e.g. favourite school subject, condition (music/silence). Bars are separated to show categories are discrete. The y-axis shows frequency or score; the x-axis shows the categories.
Histograms
Use for continuous (interval/ratio) data — e.g. reaction times, IQ scores. Bars are touching to show the variable is continuous. The x-axis shows ordered intervals (e.g. 0–9 ms, 10–19 ms…); the y-axis shows frequency. Width of bars reflects interval size.
Difference summary: bar charts have gaps; histograms don't.
Scatter diagrams
Use to display correlations between two co-variables. Each participant gives one dot; x-axis = first variable, y-axis = second. The pattern of dots shows the correlation:
- Positive: dots rise from bottom-left to top-right.
- Negative: dots fall from top-left to bottom-right.
- No correlation: dots scattered with no pattern.
A line of best fit can be added by eye, but examiners look for axis labels with units and a clear title above all else.
What to label
For any chart:
- Title above (or caption below) explaining what is plotted.
- Both axes labelled with the variable and unit.
- A scale that starts at zero where possible (otherwise note where it starts to avoid misleading).
- Bars/dots/cells legible and proportional.
⚠Common mistakes— Common errors
- Using a bar chart for continuous data (use a histogram).
- Plotting a line graph for categorical data (lines suggest continuity that doesn't exist).
- Forgetting axis units.
- Truncating the y-axis to exaggerate small differences.
AI-generated · claude-opus-4-7 · v3-deep-psychology