The Countries That Break the Pattern Are the Ones Worth Looking At
Most country indicators move together. Uzbekistan finishes school like a rich country on a fertility rate of 3.4, and Gabon has 69% internet with 20% completion.
Country indicators correlate so strongly that the informative cases are the exceptions — Uzbekistan reports 98% secondary completion on a fertility rate of 3.4, while Gabon reports 69% internet use alongside 20% completion.
Most of this site's country indicators correlate heavily with each other — under-15 share and fertility at 0.95, life expectancy and fertility at −0.83, secondary completion and fertility at −0.75.
When almost everything moves together, the ranking stops being interesting and the residuals start being interesting. Which countries sit furthest from where the pattern says they should?
Doing much better than the pattern predicts
Secondary completion against fertility rate
| Country | Fertility | Completion | Above prediction |
|---|---|---|---|
| Uzbekistan | 3.4 | 98% | +43 points |
| Samoa | 3.7 | 97% | +47 |
| Tuvalu | 3.1 | 103% | +42 |
Uzbekistan is the substantial case — a country of 37 million with a fertility rate of 3.4 reporting near-universal secondary completion. The general relationship says high fertility and high completion do not go together. Here they do.
Life expectancy against fertility rate
| Country | Fertility | Life expectancy | Above prediction |
|---|---|---|---|
| Israel | 2.7 | 83 | +11 |
| Oman | 2.4 | 81 | +7.5 |
| Palestine | 3.1 | 77 | +7.2 |
Israel is the clearest: a high-income life expectancy on a fertility rate more typical of a middle-income country. Almost every other country at 2.7 children per woman sits a decade lower.
Doing much worse
Completion against internet use
| Country | Internet | Completion | Below prediction |
|---|---|---|---|
| Gabon | 69% | 20% | −53 points |
| Equatorial Guinea | 63% | 19% | −49 |
| Suriname | 87% | 42% | −45 |
| Iraq | 80% | 45% | −37 |
Gabon and Equatorial Guinea are the two starkest rows on this page. Both report internet access in the high sixties — better than the world median — alongside secondary completion around 20%, which is close to the bottom of the entire table. Connectivity and schooling have come apart completely.
Life expectancy against fertility
| Country | Fertility | Life expectancy | Below prediction |
|---|---|---|---|
| Lesotho | 2.6 | 59 | −13.5 |
| South Africa | 2.2 | 65 | −9.6 |
| South Sudan | 3.6 | 58 | −9.2 |
| Ukraine | 1.2 | 72 | −7.9 |
What this site will and will not say about it
It will say the pattern breaks, and by how much. Every figure above is a residual from a straight-line fit across all 194 countries, and the arithmetic is reproducible from the CSV.
It will not say why. Several of these have widely cited explanations — epidemic history in southern Africa, conflict in Ukraine and South Sudan, resource economies in Gabon and Equatorial Guinea, a Soviet-era schooling system in Uzbekistan and Turkmenistan. Those are plausible and this dataset cannot test any of them. Publishing measurements rather than causes is the rule here, and outliers are exactly where the temptation to break it is strongest.
Why this is the useful way to read the data
A ranking tells you where a country sits. A residual tells you where a country sits relative to countries like it — and that second question is almost always the one being asked.
If you want to know whether a country's schooling is unusual, comparing it to the world average is nearly useless, because the world average is dominated by how rich a country is. Comparing it to the line is not.
See the underlying tables at /data, the correlations behind the line at seventeen indicators, fewer facts, or the framework that produces them at half the world is past the transition.