Data interpretation
What does a rank mean?
A rank on OurBlueDot is a comparable-sample competition rank: it orders countries that have a finite value for the same layer and period. Missing countries are unranked. Rank 1 is the highest value in that sample on current layers, not a claim of absolute truth and not a ranking of every country.
How the atlas calculates a rank
For a given layer and period, OurBlueDot collects the currently loaded country values that are finite numbers. It then assigns a competition rank: tied values share the same rank, and the next rank skips ahead by the size of the tie. Missing, null, and non-finite values receive no rank and do not occupy a place in the ordering.
On every current production layer, rank 1 is the highest value in that comparable sample. The rank direction is not reversed to mean “best” or “least harmful.” Adult obesity rate in the United States therefore uses rank 1 as the highest age-standardized prevalence, not the least healthy country. Road traffic deaths in Sweden uses rank 1 as the highest death rate among countries with a value.
Country pages spell this out as: among the countries with comparable values, this country ranks N. They also state that the rank describes the available comparable sample and is not a complete ranking of every country. Some layers add a specific caveat. Employment Rate notes that a higher rank does not mean a better or more prosperous labour market. Adult Obesity Rate states that a higher rank does not mean a less healthy country.
The comparable sample is not the world
The sample is the set of countries with a finite value for that period, not the 193 UN members and not every territory in the source workbook. If a country is grey, it is unranked. If the source sample is small, as on some mobility layers, rank 1 is rank 1 among those observations only.
A rank is withheld when fewer than 2 comparable countries are available. The comparable-sample median is quoted only when at least 3 finite values exist. Employment rate in Finland is therefore ranked among other ILO modelled employment-to-population ratios for that year, not against unemployment rates, and not against countries the source does not estimate.
Rank is not absolute truth
A rank is an ordering of the numbers the atlas currently holds. It moves when the comparable set moves: another reporting year, a source revision, or a country entering or leaving the finite sample. It does not adjudicate living standards, health systems, or policy.
Happiness is the explicit teaching case for an external rank. Happiness in Finland stores the World Happiness Report published world rank separately. The number shown as the map rank is the comparable-sample rank among currently loaded countries, not the report’s published rank. The two can differ.
This does not measure
A comparable-sample rank is not a complete league table of every country, not a quality-of-life index, and not the same thing as a source’s own published ranking. It does not convert a descriptive indicator into a verdict. Health and Economy layers keep their own definitions; ranks are not blended across them.
Continue in the atlas
These pages are the working examples for this question. They are not a complete catalogue.
- Methodology
- Sources
- Health theme
- Economy theme
- Adult Obesity Rate
- Employment Rate
- Happiness
- Road traffic deaths
- Adult Obesity Rate in United States
- Employment Rate in Finland
- Happiness in Finland
- Road traffic deaths in Sweden
Related atlas concepts
Other Insights pages that explain shared atlas behaviour, not a reading list of articles.