The zero that was never measured
When a chart expects a number, software often fills an empty field with zero. That is convenient for computation and dangerous for interpretation. Zero homicide means a reported rate of zero. Missing homicide data means the system does not have a usable observation. Those are different statements.
The same rule applies to child labour, suicide mortality, freedom of expression and carbon emissions. A blank field can be caused by survey timing, reporting capacity, conflict, changes in classification or the absence of a comparable national series.
Four kinds of gaps
| Gap | Example | Correct response |
|---|---|---|
| Absent | No national value in the connected series | Show “not available” |
| Stale | Latest value is many years old | Show year and restrict interpretation |
| Partial | Only some countries contribute to a regional average | Publish coverage beside the average |
| Modelled | Estimate combines reported data and statistical modelling | Name the producing institution and method |
Why missingness is socially patterned
Data gaps are not random. Countries facing conflict, displacement, weak civil registration or limited statistical budgets may also face greater human pressure. Excluding them without explanation can make a regional average look safer than the region actually is. Conversely, assuming that every missing value is high would also be unjustified.
The responsible answer is not to guess. It is to show the boundary of the evidence and explain how the missing observations could affect interpretation.
Suicide data shows the stakes
WHO states that only about 80 Member States have good-quality vital-registration data usable directly for suicide estimates. Stigma and misclassification can add further uncertainty. A cross-country table that ignores those differences risks rewarding strong measurement systems with apparently worse numbers while undercounting harm where reporting is weaker.
Regional averages need a denominator
“Four countries reported” is not enough information without knowing whether the region contains five countries or fifty-four. Humain Alert now writes coverage as a fraction and marks averages with weak coverage as unsuitable for a regional conclusion. No data is better than a false sense of precision.
The display rule
- Use an em dash or “not available,” never a numeric zero, for absence.
- Keep the last known year next to stale observations.
- Withhold a gauge when the reference year or geographic coverage is insufficient.
- Describe imputation explicitly if it is ever introduced.
- Preserve the source link so readers can inspect the original series.
The absence of a number is information about the measurement system—not evidence that the underlying problem is absent.
Primary sources and data
- WHO — Suicide fact sheet ↗Includes data-quality limitations
- World Bank — DataBank metadata ↗Definitions, sources and coverage
- UNICEF Data — Child Labour Global Estimates 2024 ↗Survey and methodological materials
Evidence note: Sources were checked on 7 Sep 2026. Publication years differ because each institution follows its own release schedule. Humain Alert preserves those dates instead of presenting the series as simultaneous.