Genderyzer is an experimental research tool for determining the gender mix of a list of names by Jofish Kaye. Helpful feedback welcomed.
Genderyzer takes as input a text file with one name per line, in the order First Initials Last, ignores everything except First, and returns statistics based on the gender mix of the names, best as it can. (Blank lines and those starting with # are ignored.) You probably want it in UTF-8, although it'll try to do the right thing with HTML-encoded characters. Only the most recent file uploaded is kept on the server; it is automatically overwritten whenever a new file is uploaded.
Gender information for names is taken from a variety of sources, including census records, government name guides, and baby name lists. Please let me know if you feel I have incorporated information from your lists and you'd like it removed.
All results are approximate. No guarantees. Please reference this site if you use this service. If you publish academic papers that use this work, please reference this paper: http://doi.acm.org/10.1145/1520340.1520364
15 Oct 09: I recently discovered a design flaw where 29 names were being incorrectly reported. See 29names.txt for details. My apologies; please do continue to send feedback if you notice errors.
This page design adapted from RedLine by Igor Jovic.
| Male | 0 | 0% |
| Female | 20 | 5% |
| Initials | 5 | 1% |
| Unknown | 375 | 94% |
| Ambiguous | 0 | 0% |
| Total | 400 | 100% |
Download annotated version of most recently uploaded file here