To put statistics in context, explain what is being counted, how large the relevant group is, what comparison matters, which summary is being used, and how much uncertainty remains. Pair the number with a plain-English explanation of its significance—and its limits—rather than expecting digits to do all the talking.

Consider this announcement: “Complaints increased by 100%.”

Time to panic? Perhaps. But if complaints rose from one to two while orders increased from 100 to 1,000, the story looks different. Complaints doubled, but the complaint rate fell from 1% to 0.2%.

Nothing about the original statement was mathematically wrong. It was simply missing the information needed to understand it.

That is the trick to sounding smarter with statistics: not collecting impressive numbers, but making their meaning clear. Here are five ways to do it. All numerical scenarios below are hypothetical.

1. Ask “Out of How Many?”

A count tells you how many times something happened. A rate or proportion relates that count to another quantity, such as the number of residents, customers, or transactions. That denominator—the “out of how many?” part—is essential context.

Imagine two delivery companies reporting damaged packages during the same month:

| Company | Damaged packages | Total packages | Damage rate | |---|---:|---:|---:| | Parcel Panda | 40 | 20,000 | 0.2% | | Box Badger | 15 | 3,000 | 0.5% |

Parcel Panda damaged more packages in total. Box Badger had the higher damage rate.

Both facts matter, but they answer different questions. A customer comparing these figures might care most about the rate. A manager planning replacement shipments also needs the total.

The Australian Bureau of Statistics’ guidance on communicating statistical findings recommends using rates when comparing populations of different sizes. It also stresses including the context and definitions needed to interpret a result.

Before repeating a number, ask:

  • Out of how many people, purchases, or events?
  • Over what period?
  • Among which group?

Try saying: “There were more damaged packages overall, but a smaller proportion of deliveries were damaged.”

That is more informative than declaring one company “worse” and leaving the denominator hiding behind the sofa.

2. Give Percentage Changes a Starting Point

“Returns increased by 50%” sounds substantial. But 50% of what?

Suppose an online store’s return rate rises from 2% to 3%. There are two accurate ways to describe the change:

  • Relative increase: 50%, because the increase of 1 is half the starting value of 2.
  • Percentage-point increase: 1 percentage point, because 3% minus 2% equals 1 percentage point.

Percentage change and percentage-point change are different measurements. Confusing them can make a modest-looking shift sound enormous—or make a substantial relative change disappear.

For the hypothetical store, translate the percentages into equal-sized groups:

“The return rate rose from 2% to 3%—equivalent to 20 versus 30 returns per 1,000 orders.”

Now readers can see both the starting point and the difference.

The calculations are straightforward:

Relative percentage change = (new value − original value) ÷ original value × 100

Percentage-point change = new percentage − original percentage

Neither description is automatically more honest. They answer different questions. The strongest explanation often gives the before-and-after figures, allowing people to judge the scale themselves.

Also, watch for tiny starting values. Going from one late shipment to three is a 200% increase. That does not make the increase imaginary; it makes the original count important.

Try saying: “That is a 50% relative increase, from 2% to 3%.”

The extra words are doing useful work.

3. Choose a Comparison That Actually Fits

A statistic becomes more useful when you compare it with something relevant: an earlier period, a target, a similar group, or a familiar unit. The comparison should clarify the finding without changing what is being measured.

Suppose a community center’s electricity bill falls from $24,000 to $18,000 per year.

You could say:

“The center saved $6,000 annually, or an average of $500 per month.”

That gives the annual total an everyday scale. It does not mean every month’s bill fell by exactly $500.

Now consider the claim “Attendance increased by 20%.” Compared with last month? The same month last year? A week when the building was closed for repairs?

Before celebrating, check that the comparison uses matching definitions, time periods, and units. Presenting the complete picture, including relevant limitations, is a central principle of statistical communication.

Useful questions include:

  • Are both figures monthly totals, or is one annual?
  • Does “attendance” mean visits or unique visitors?
  • Were opening hours comparable?
  • Was the starting period unusual?

For unfamiliar quantities, an everyday comparison can help. Uncover Words’ guide to explaining complex ideas with familiar comparisons offers a useful approach: explain the match and acknowledge its limits.

Try saying: “Attendance was 20% higher than in the same month last year; both periods had the same number of opening days.”

Now the comparison has a foundation, not just a confident expression.

4. Find Out What “Average” Means

“Average” sounds reassuringly specific. Unfortunately, it can arrive wearing several different hats.

Two common measures are:

  • Mean: Add all the values and divide by how many there are.
  • Median: Arrange the values in order and find the middle; with an even number of values, average the middle two.

The mean can be strongly influenced by unusually high or low values. The median is less sensitive to those extremes, which is why it can be useful when describing a typical income. The UK’s Office for National Statistics explains this distinction in its discussion of mean and median household income.

Imagine five annual salaries:

$40,000, $45,000, $50,000, $55,000, and $310,000.

The mean salary is $100,000. The median is $50,000.

Both are correct. But “the average employee earns $100,000” could leave listeners with a misleading impression of what most employees earn.

Also ask how spread out the values are. In a hypothetical café, waits of 9, 10, and 11 minutes have the same mean as waits of 1, 2, and 27 minutes: 10 minutes. The customer experience is hardly identical.

Try saying: “The median salary is $50,000; the mean is higher because of one unusually large salary.”

You have explained the number without making anyone sit through an unexpected statistics lecture.

5. Explain Where the Number Came From—and What It Cannot Prove

A statistic needs a source, not just a percentage sign.

If a survey says “70% prefer the new design,” ask who answered, how they were recruited, and what they were asked. Question wording and the order of questions can affect responses.

Then check uncertainty. A small difference between survey results may not justify a confident declaration that opinion has changed. Pew Research Center’s tips for interpreting polls caution against overstating precision and explain that a margin of sampling error does not cover every possible source of survey error.

Finally, separate a pattern from an explanation.

Suppose people who attend a voluntary training session subsequently earn higher performance scores. Did the training help? Possibly. But perhaps particularly motivated employees were more likely to attend.

An association alone does not establish cause and effect; other factors may explain the relationship.

Try saying: “Employees who attended scored higher afterward, but these figures alone do not show that training caused the difference.”

When details are missing, use specific questions to ask for clarification: “Who was included?” is more useful than a vaguely suspicious “Really?”

Make the Meaning Easier to See

Before sharing a statistic, finish this sentence:

“This number means ___, compared with ___, among ___, over ___.”

Then add any limitation that would materially change the interpretation.

You do not need to attach a research appendix to every conversation. Usually, one well-chosen detail—a denominator, starting value, comparison, or caveat—can make the explanation clearer.

The goal is not to make numbers sound bigger, smaller, or more impressive. It is to help people understand what they actually say.

That is a much better way to sound smart than announcing “200% growth” and hoping nobody asks what you started with.

Share: