The dataset
Every figure in the game, as a file you can download, licensed CC BY 4.0. 24 countries, 23 of them backed by a traceable primary study covering 35,830 measured men, 16 with a direct link to the paper.
It is published because the alternative is worse. The figures that circulate about this subject come overwhelmingly from a world map that has been fabricated for fifteen years and copied ever since. This is the same information with its provenance attached.
Download
- penis-size-by-country.csv — one row per country, UTF-8, RFC 4180 quoting.
- penis-size-by-country.json — the same rows as an array of objects.
- datapackage.json — a Frictionless Data descriptor with the field types and the licence, for anything that reads packages rather than files.
These are generated on request rather than committed, so a download is always current and never contains an answer the game has not given up yet — see below.
What the columns mean
| Column | What it holds |
|---|---|
| country_code | ISO 3166-1 alpha-2 country code. |
| country_name | Country name in English. |
| stretched_length_cm | Mean stretched penile length in centimetres, exactly as published. Empty when withheld or untraced. |
| sample_size | Number of men measured in the cited study. 0 where no primary study was traced. |
| measurement_method | How the study measured, e.g. whether the ruler was pressed to the pubic bone. |
| source | Study citation: authors and journal. |
| source_year | Year of publication. |
| source_url | Link to the paper, where one was verified to resolve to it. Empty otherwise. |
| withheld | True while the country sits inside the game's three-day answer-suppression window. Re-download later for the figure. |
| untraced | True when a figure circulates widely but no primary study could be found. Such rows carry no length. |
Two things it deliberately does not do
It does not mix measurement methods. Every row is a stretched length — average penis size — because that is the figure a clinician takes with a ruler. Countries whose only published figure is an erect measurement are absent rather than converted, which is why the table is 24 rows and not 190. Relaxing that would roughly double the coverage and make every comparison in it meaningless. The rejected candidates are named.
It does not quietly fill gaps. A row with no traceable primary study carries untraced = true, an empty length and a sample size of zero, rather than borrowing a number from an aggregator. Rows are withheld, never invented.
Why some rows are blank today
The game withholds a country's figure for the three days around its turn in the rotation, and the download honours that: those rows keep their provenance and carry withheld = true with an empty length. Otherwise a download would be a spoiler service.
It is a live export rather than a file, so re-downloading after the window has passed gives the complete table. At any given moment at most three of 24 rows are affected.
Licence and citation
Licensed Creative Commons Attribution 4.0. Use it commercially, redistribute it, build on it; the only condition is attribution, which is the entire point of publishing.
To be precise about what is licensed: the figures are facts taken from published papers, and a mean is not copyrightable. What this covers is the compilation — the choice of metric, the inclusion and exclusion decisions, the provenance fields and the written notes. If you use a single country's figure, cite the study named in its row, not this page.
Cite as
Dickle. Average stretched penile length by country [Data set]. Version 1.0.0. https://www.dickle.online/data
@dataset{penis_size_by_country,
title = {Average stretched penile length by country},
author = {{Dickle}},
year = {2026},
version = {1.0.0},
url = {https://www.dickle.online/data},
license = {CC BY 4.0}
}Corrections
If a row is wrong, or you can point at a primary study for a country that has none, that is the most useful thing anyone can send us. Get in touch, and see the methodology for how each figure was chosen and the four ways it can be wrong.