Reading path: Marine organisms and body plans →
Start with the claim
A statistic such as “shorebird numbers rose by 20 percent” sounds precise, but it leaves important questions open. Which shorebirds? At what coast or set of sites? Compared with which year? Were observers counting individuals, nests, or sightings? Until those details are known, the percentage cannot tell a reader whether the change is broad, local, or even based on comparable observations.
First locate the original report or dataset rather than relying on a headline, infographic, or repost. Record who collected the information and whether the figure is a direct count, a model estimate, or a summary of several studies. Secondary sources can be useful guides, but they may shorten a method or omit a qualification that changes the interpretation.
Check period, units, and denominator
The date range sets the scale of the claim. A comparison between two breeding seasons answers a different question from a trend across several decades. Seasonal migration, storms, and survey timing can shift what is observed, so ask whether the compared periods cover similar parts of the year and use the same sites.
Units and denominators show what the number actually represents. A count per kilometer of shoreline is not equivalent to a total coast-wide count. A percentage of sampled sites with debris is not the percentage of all beaches affected. For example, if four of ten surveyed coves contain a particular item, the result describes those sampled coves; it does not automatically establish that 40 percent of every cove in the region contains it.
Read how the evidence was made
Look for the sampling design, definitions, and any changes in equipment or observer practice. A new camera, a different search area, or a revised species-identification guide can alter counts even if the underlying ecosystem has not changed. A useful report explains the effort behind the result: how many surveys were made, where, and under what conditions.
Uncertainty also matters. A modeled estimate may be reported with a range, while a field count may be affected by imperfect detection. These are not flaws to hide; they describe how strongly the evidence supports a conclusion. Compare like with like, and avoid presenting a small sample as a complete inventory.
Repeat the statistic responsibly
When sharing a number, preserve its scope and qualification in the same sentence. Instead of saying “the population increased,” say that a specified survey recorded an increase at its monitored sites over the stated interval. Add the source and date so another reader can inspect the method.
A practical check is to write down the measured thing, place, period, unit, denominator, and method before deciding what the result means. If one field is missing, describe the gap rather than filling it with an assumption. That habit turns a striking figure into evidence that can be examined, compared, and used with care.
Have a correction or a source to suggest? Send a note →