THE SHORT ANSWER
Check the base numbers, denominator, time period, population, sample selection, definition, uncertainty and visual scale. Distinguish relative from absolute change and association from causation. Missing context can mislead even when the arithmetic is correct.
Ask what the number is a number of
| Presentation | Missing question |
|---|---|
| Conversion improved 50% | From what base rate and over what volume? |
| The average order is ₹2,000 | Mean, median or another average; how spread out? |
| Eight out of ten prefer it | How were ten people selected and asked? |
| Complaints doubled | From how many, while customers changed by how much? |
| The chart rises sharply | Where does the axis start and which dates were selected? |
Relative change can magnify a small baseline
If a rate rises from 2% to 3%, the relative increase is 50% and the absolute increase is one percentage point. Both statements are mathematically correct; together they communicate more than either alone.
A missing denominator has the same problem. Twenty failures could be alarming in 100 attempts and less informative in ten million.
The sample defines what can be concluded
A poll of active customers cannot directly describe people who abandoned the service. A success story selected after the result cannot estimate the success rate. Ask who had a chance to appear, who is absent and whether definitions changed.
Clear statistical communication should surface context, quality limitations and uncertainty rather than hide them behind precision.
Evidence & context: Office for Statistics Regulation
Skepticism should improve interpretation
- Reconstruct the base numbers.
- Check population, period and definition.
- Inspect the axis and omitted range.
- Look for uncertainty and sample size.
- Separate description from causal explanation.
- Ask whether another reasonable summary changes the decision.
Sources & further reading
- Short guide to Code of Practice Standard Nine: Be Clear
Office for Statistics Regulation. Official guidance on presenting statistics with context, limitations and uncertainty. It does not make every statistical interpretation equally valid.
- Reproducibility and Replicability in Science
National Academies of Sciences, Engineering, and Medicine. A consensus report on evidence, uncertainty, transparency, reproducibility and replication. Scientific standards need proportionate adaptation outside research settings.
Examples and exercises are illustrative unless attributed to a source. No independent expert review is claimed.
A correction, a counterexample or an experience worth sharing?
Join the conversation ↗