3

Wikipedia has articles about given names (example: John).

In each of these articles, I would like to add a histogram showing how popular the name has been through the ages. For each year, it could be the number or proportion of children who received this name.

Where can I find data for this?

Requirements:

  • Must be reusable in Wikipedia (CC-BY-SA 3.0)
  • Since as far back as possible, for instance there have been Johns since 134BC at least. Estimation data is OK.
  • Some will argue that John was actually spelt another way and should be considered another name. Any stance is fine, as long as the data stays consistent with the stance they have chosen.
  • I am looking for data for the whole world, but dataset for parts of it (for instance Chinese names only) are welcome too.
  • 1
    What do you mean by popularity? Naming a child or using the name in literature? Here's one popular source for naming: ssa.gov/oact/babynames . As for mentions, Google Ngrams are still the most comprehensive: books.google.com/ngrams . – Anton Tarasenko Jun 10 '15 at 11:49
  • I've seen lots of tools / visualizations out there, but they always seem to be US centric, and limited to the time range of the SSA data that Anton mentioned. – Joe Jun 10 '15 at 15:18
  • @Anton: Number of children who are given this name each year. – Nicolas Raoul Jun 11 '15 at 3:38
  • @NicolasRaoul : the SSA website has that information for the US. Click the 'Popular Names by Birth Year' tab, and then click 'Number of births' – Joe Jun 11 '15 at 18:54
5

Here You have names with the popularity in capital city of Poland: Popularne imiona w Warszawie

  • Thanks! So apparently it only goes back to 2004? Or maybe I am not looking at the right place? – Nicolas Raoul Jun 12 '15 at 2:30
5

For data beyond the SSA dataset (that goes back to 1880, but has full coverage only back to 1932), you may look at this site

http://www.galbithink.org/names/agnames.htm

providing names for US census data and from British sources. Note that some data are normalised (fortunately the normalisation patterns are Open Source, so you can inspect them and make up your judgement).

5

For the USA, the Social Security Administration has statistics going back to 1880.

The whole national data is downloadable in a ZIP file available for download at http://www.ssa.gov/OACT/babynames/limits.html
Direct link: http://www.ssa.gov/OACT/babynames/names.zip

Excerpt:

yob1880.txt:

Mary,F,7065
Anna,F,2604
Emma,F,2003
Elizabeth,F,1939
Minnie,F,1746

License: Public domain per http://www.ssa.gov/agency/websitepolicies.html#a0=7

  • @jknappen: Thanks for the link! Believe it or not, I still can't find the page for the 400-1000 interval... – Nicolas Raoul Jun 15 '15 at 9:42
  • Indeed it contains ALL data even though the page says "Beyond the Top 1000 Names", thanks! – Nicolas Raoul Jun 15 '15 at 9:56
5

For the USA there is another great source on names: The Social Security Death Index (SSDI). Especially it fills the sparseness in the SSA data predating 1932.

As the name tells, it contains only data of deceased people, not of the ones still alive.

EDIT: The Social Security Master Death File (SSDMF) is available from http://ssdmf.info/ It contains Names, Day of Birth and Day of Death. Data go back to 1800 (but the oldest points are probably input errors).

P.S. Note that only the first 10 letters of the given name are recorded in the SSDMF: You find no CHRISOTPHER, but lots of CHRISTOPHE's.

  • Great! Where can it be downloaded? What year does it go back to? Deaths can be correlated to births, at least as a good enough approximation to fill the voids. Thanks! – Nicolas Raoul Jun 16 '15 at 2:46
  • @Nicolas Raoul See EDIT – jknappen - Reinstate Monica Jun 16 '15 at 7:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.