What is the likelihood of having a biography on Wikipedia? How does this likelihood vary across the 10 major largest editions of Wikipedia? Does country of birth affect the probability of having a biography? Which countries are overrepresented, and do different editions rank them differently? By combining demographic birth data from the United Nations World Population Prospects with biography counts by year and country of birth, this study offers new insights into geographical bias on Wikipedia.

Looking at raw biography counts in English Wikipedia, the United States leads with 337,763 biographies, followed by the United Kingdom (158,718) and Germany (69,204). India ranks 8th with 35,084 biographies and the China 14th with 18,648.^[1] These figures, however, reflect population size and historical documentation practices as much as anything else — to draw meaningful comparisons, they must be weighed against demographic data.

In 2022, I conducted an initial study comparing birth counts by department in metropolitan France, as provided by the French statistical office, with the number of biographies by department of birth.^[2] That was my first attempt to measure the likelihood of having a Wikipedia biography, though the scope was limited to France. I've been looking for a worldwide birth dataset to extend my analysis.

The United Nations World Population Prospects (WPP) turned out to be the perfect dataset for this purpose: it provides birth estimates by year and country going back to 1950, enabling a global measure of the likelihood of having a Wikipedia biography.

I've published a first analysis for the French-language edition of Wikipedia.^[3] My analysis shows that very small and rich countries such San Marino, Monaco, and Andorra lead the ranking with more than 1 biography per 1,000 births and that and that major Asian countries such as People's Republic of China and India are at the very bottom of the rankings, with less than 1 per 100,000.

In this analysis, I propose to measure likelihood of having a biography on a global scale and at the country level for 18 versions of Wikipedia. To my knowledge, this is the first study which computes the likelihood of having a biography at the country level.

For the purpose of this study, I've collected the Biographies vs Births dataset, a dataset which combines birth data from the World Population Prospect and Wikipedia biography counts from Wikidata.^[4]

More precisely, I use the estimated number of births (Births) to get number of births by country from 1950 to 2019 from 2024 edition of the World Population Prospects.^[5] Rather than tracking historical borders and former countries, the WPP uses current country definitions, which simplifies long-term cross-country comparisons.

I've collected number of biographies by country, and year of birth between 1950 and 2019 using Wikidata SPARQL queries.^[6] Data were collected on July 4, 2026. 18 editions of Wikipedia are included in the dataset: Arabic (ar), Czech (cs), German (de), English (en), Spanish (es), Persian (fa), French (fr), Hebrew (he), Italian (it), Japanese (ja), Korean (ko), Dutch (nl), Polish (pl), Russian (ru), Turkish (tr), Ukrainian (uk), Vietnamese (vi), and Chinese (zh) Wikipedia.

It is useful to keep in mind that the data could have some drawbacks. Some Wikipedia articles are not connected to a Wikidata item and some Wikidata items don't have precise information about birthdate and birthplace.

I estimate the likelihood or probability of having a biography in Wikipedia using the ratio of number of biographies and number of births over a time period. To keep the number of decimal places manageable, I multiply the ratio by one million. This gives the biographies per million births (bmb) ratio.

In this study, I focus on people born between 1950 and 1989. People born after 1990 have had less time to meet Wikipedia's notability criteria and are therefore systematically underrepresented in biography counts.

The biography per million births ratio is calculated both globally for each wiki and at the country/wiki granularity.

In this section, I first look at biographies per million births ratio at the global level, then I focus on the English Wikipedia, and finally I look at similarities and differences between the 18 editions of Wikipedia included in the study.

All results are produced in an Observable notebook.^[7]

Between 1950 and 1989, the World Population Prospects records a total of 4.7 billion births (4,726,859,446 births) across 216 countries. The biographies per million births ratio ranges from 5.5 for Vietnamese Wikipedia to 113.3 for the English edition. Nothing surprising here, this is just due to the difference in the number of biographical articles in the different editions of Wikipedia but this is useful to give some order of magnitudes of the probability of having a biography.

In this section, we focus on the English Wikipedia.

Globally, the results reveal stark disparities between countries: the likelihood of having a biography ranges from 7.5 per million births in Ethiopia to 13,340 in San Marino — meaning that while someone born in Ethiopia has roughly a 1-in-130,000 chance of being notable enough to warrant a biography, someone born in San Marino has closer to a 1-in-75 chance. People born in San Marino are almost 2,000 times more likely of having a biography than those born in Ethiopia.

The five countries with the highest likelihood of having a biography in English Wikipedia are San Marino (13,339.6 biographies per million births), Andorra (8,593.8), Liechtenstein (7,922.5), Falkland Islands (6,082.7), and Iceland (5,302.5). The first four are very small countries with fewer than 15,000 births recorded over the 40-year period. Browsing the list of biographies by country show that local athletes could explain the relatively high number of biographies for those countries.^[8]

Globally, the top 20 includes a mix of very small countries such as Monaco, Faroe Islands, Cook Islands, Tonga, Cayman Islands, Antigua and Barbuda, Isle of Man, and Greenland, and small or relatively small countries such as Iceland, Luxembourg, Estonia, Norway, New Zealand, Ireland, and Sweden.

There is a decreasing relationship association between a country's demographic size, measured by the number of births between 1950 and 1990, and the probability of having a biography. Overall, a 1% increase in the number of births is associated with a 0.407% decrease of the probability of having a biography.

Figure 1 shows that English speaking countries such as New Zealand, Ireland, Canada, Great Britain, and the United States of America are far above the regression line, which means that their biographies per million births ratio is high relative to their birth count.

Figure 2 shows regional disparities of the biographies per million births ratio.

Western nations have a all a high (above the 100) or very high (above 1,000) value of biographies per million births ratio: the United Kingdom have 1,361.3 biographies per million births, Canada 1,265.4, the United States (753.4), France (521.4), and Spain (482.8).

Latin American countries sit in the middle of the ranking. Argentina (279 biographies per million births) and Chile (232) exceed the global average, while Mexico (84), Brazil (80), and Peru (63) fall in the 50–100 range — well below the regional leaders but far from the bottom of the distribution.

Some Asian countries perform well above the regional average, including Japan (220 biographies per million births), South Korea (153), Malaysia (139), Taiwan (108), and Kazakhstan (102), all exceeding 100 biographies per million births. Many of the continent's most populous countries, however, rank near the bottom of the global distribution: India records just 16 biographies per million births, Indonesia 15, Myanmar 13, Vietnam 13, North Korea 13, Yemen 12, Bangladesh 10, and China 8.

Among large African countries, South Africa is the best ranked with 145.4 biographies per million birth. Other African countries are below 100 biographies per million births. Ethiopia is the last one in the English Wikipedia with only 7.5 biographies per million births.

The biographies per million births ratio is highly correlated across the 18 editions of Wikipedia included in the studies. Overall, countries with the high biographies per million births ratio in one edition have a high biographies per million births ratio in another edition. All coefficient of correlation are greater than 0.7.

Since the correlation between the 18 editions of Wikipedia included in the study is very high, countries at the top of the ranking are often the same. San Marino is 1st in the English, German, French, Italian, Chinese and Polish editions, 2nd in the Spanish one, 7th in the Dutch one, 8th in the Japanese one, and 9th in the Persian edition. Andorra, Liechtenstein, Iceland, and Monaco are always in the top 10 and Nauru is in the top 10 9 times out of 10.

Though the rankings are globally similar across the 18 Wikipedia editions, there are also some notable differences.

In the Japanese edition, several Asian and Pacific countries rank notably higher than in the English edition: Japan climbs from 103rd to 3rd, the Republic of Korea from 116th to 54th, Taiwan from 129th to 77th, and the Democratic People's Republic of Korea from 205th to 134th. Pacific island nations also fare better, with Tuvalu rising from 19rd to 4th, Tonga from 15th to 5th, and the Marshall Islands from 86st to 14th. Overall, Asian and Pacific countries are better represented in the Japanese edition than in the English one.

We find the same pattern for the Chinese edition: Taiwan climbs from 129nd to 12th, China from 211th to 124th, the Republic of Korea from 116th to 43rd, Japan from 103th to 26th, and the Democratic People's Republic of Korea from 205th to 129th. Tuvalu rising from 19th to 4th, and the Marshall Islands from 86th to 14th.

In the Persian edition, some Middle-East countries are better represented than in the English edition: Qatar goes from 29th rank to 4th, Armenia from 91th to 6th, United Arab Emirates from 50th to 8th, Bahrain from 75th to 16th, Kuwait from 97th to 33th, and the Islamic Republic of Iran from 149th to 48th.

In the Dutch edition, Tuvalu jumps from 23th to 1st position, the Netherlands from 48th to 2nd, Belgium from 55th to 3rd, Suriname from 62th to 5th, and Curaçao from 59th to 13th.

The Polish edition is very similar to the English edition. The main difference is the ranking of Poland (from 93th to 10th).

The Italian edition is also very close to the English one. Italy is ranked 8th instead of 78th.

In the French edition, countries with a French-speaking community are noticeably better represented. Switzerland rises from 35th to 10th, France from 68th to 14th, and Belgium from 55th to 13th. The effect extends beyond Europe: Gabon climbs from 113th to 68th, Tunisia from 130th to 76th, and Canada from 25th to 17th.

In the Spanish edition, Spanish-speaking countries are better represented than in the English one. Falkland Islands is 1st vs 4th, Spain 9th vs 72th, Uruguay 10th vs 54th, Argentina 15th vs 95th, Chile 18th vs 101th, etc.

For the German edition, we also find a bonus for German-speaking countries such as Germany (12th vs 71th), Switzerland (11th vs 35th) and Austria (8th vs 42th). One notable difference is Timor-Leste which jumps from 127th position in the English edition to the 44th position in the German one.

This study demonstrates that combining demographic data with biography counts offers a new perspective on the geographic biases of Wikipedia. It brings new perspectives in the field of notability studies or representation studies of Wikipedia.

The gap between the top and bottom of the distribution is striking, and the results appear highly consistent across the 18 Wikipedia editions included in the study.

A cluster of very small countries — San Marino, Monaco, Andorra, Liechtenstein, and Nauru — consistently rank at the top of the distribution, while populous Asian countries such as India and China remain at the bottom.

In addition to this study, I release the first version of the Biographies vs Births dataset. Researchers, journalists and Wikimedians can go further by looking at probability of having a biography though Wikipedia edition, country, year of birth and gender.

In the future, the dataset could be extended to a larger number of Wikipedia editions and to other Wikimedia projects such as Wikiquote, Wikisource and Wikidata.

Further research would be valuable to understand what drives these disparities. In particular, it would be worth investigating whether the gap stems from differing likelihoods of meeting the notability criteria, or from how those criteria are interpreted across Wikipedian communities.