HowHint
Original editorial illustration; not a location photograph or product screenshot.
A chart can be numerically correct yet visually deceptive. A dramatic slope may result from a narrow axis; two impressive percentages may refer to different denominators. Learning to read visual evidence starts before interpreting the trend: understand what was measured, how it was drawn, and what was left out.
Identify what each mark represents
Begin with title, axes, legend, and units. Is the horizontal axis time, location, ranking, or a numeric measurement? Is the vertical axis raw counts, rates, percentages, an index, or a logarithmic scale? If a chart compares countries, ask whether population size or age structure should be considered. The honest visualization of one quantity is not automatically the right comparison for another question.
Examine baselines and intervals
A bar chart with a truncated vertical axis can magnify small differences in bar height. For line charts, a truncated axis may be reasonable when describing small changes, provided the scale is visible. Check whether time intervals are evenly spaced; a line drawn across missing months can imply observation where there was none. A chart that begins at an unusually convenient year may hide the longer pattern.
Translate rates back into numbers
Suppose an event rate increases from 1% to 2%. That is an increase of one percentage point and a 100% relative increase. Both statements can be mathematically correct and emotionally different. Ask for the baseline and denominator before concluding whether the change is large in practical terms.
Look for uncertainty and selection
Survey results may include confidence intervals, sample sizes, and subgroup limitations. Observational relationships do not prove causation. A chart that shows only the most successful companies, provinces, or test cases may be affected by selection bias. Check for missing categories and verify whether the source labels estimates differently from measured data.
Reconstruct one example
Pick two adjacent values from a chart and write them in a plain sentence. Then calculate the absolute difference and the relative difference with the correct base. If you cannot reconstruct the numbers from the supplied chart or linked dataset, avoid treating the graphic as self-explanatory. Search for the original methodology or data table.
Ethical chart sharing
When reusing a graphic, preserve its axis labels, units, source attribution, and caveats. If you crop a figure to fit a phone screen, ensure the crop does not remove information essential to the inference. Prefer a clear descriptive alt-text that states the trend and the range, not an exaggerated headline.
FAQ: Is a zero baseline always required? Not for every chart type. It matters especially when bar lengths encode magnitude. The correct rule is to avoid a visual impression inconsistent with the data.
A worked percentage comparison
Use this original planning worksheet with your own information. Its example decisions illustrate a method; they are not results from a product test or a visit.
| Check | What to record | How to use it |
|---|---|---|
| Starting rate | 1 in 100, or 1% | Write the denominator as well as the percentage. |
| Ending rate | 2 in 100, or 2% | The absolute increase is 1 percentage point. |
| Relative change | (2 − 1) ÷ 1 × 100 | The relative increase is 100%; neither description alone gives full context. |
Download this planning worksheet (CSV)
Sources & verification
Source check: 2026-10-28. This is desk research and practical guidance; no firsthand product test or place visit is claimed. Provider terms can change.
