A number can be correct while the conclusion attached to it is misleading. “It doubled” describes a proportional change, but gives no starting value. An increase from two to four and an increase from two thousand to four thousand need different practical responses.
Before sharing a statistic, ask what was counted, out of how many possible cases, where and during which period. Keep those details beside the number. Without them, a percentage is an incomplete description.
Test a comparison with small numbers
Imagine one learning group records four missed sessions among twenty scheduled sessions. Another records six among sixty. The second group has more missed sessions in total, but the first has the higher proportion: 20 per cent compared with 10 per cent.
Neither calculation decides which comparison is useful. Are you planning replacement sessions, comparing attendance patterns or investigating barriers to participation? The question determines which measure matters.
Next, check whether the groups counted in the same way. Were sessions the same length? Were cancellations included? Did the figures cover comparable periods? A neat calculation cannot repair incompatible definitions.
Choose a statistic from an article and trace it to the underlying source. Record the population, measure and period. Compare the source's wording with the headline and note anything you could not establish.
Gapminder's Factfulness resources examine recurring ways people misinterpret information. A useful habit is to keep the context attached to a number and narrow a claim when the evidence supports less than the headline suggests. That habit matters just as much when a statistic agrees with your expectations.