?

Log in

Months, Seasons, and Names - Office of Naval Contemplation

Jan. 6th, 2012

12:58 pm - Months, Seasons, and Names

Previous Entry Share Next Entry

The other day, I posted some vague musings on G+ on the haphazard appropriation of month and season names as personal names. In the names of Science!, I've decided to do some actual data analysis rather than just going off of anecdotal impressions and recollections. I'm looking at the Census names files (a compilation of the frequency of each personal name that shows up in at least 0.001% of the sample for each gender in the 1990 Census post-enumeration survey) and extrapolating the total number of people with each name in the US, based on the most recent Census bureau estimate of 148.5 million men and 152.9 million women total, and rounded to the nearest 500 to avoid implying too much false precision (the precision of my data is 0.001% of the sample, or about 1500).

NameMenWomen
Winter01,500
Spring03,000
Summer026,000
Autumn026,000
Fall00
January01500
February00
March00
April0235,500
May044,500
June0191,000
July00
August22,000 (*)0 (**)
September01,500
October00
November00
December00
Sunday01,500
Monday00
Tuesday00
Wednesday00
Thursday00
Friday00
Saturday00

(*) Plus 10,500 men named Augustine and 7,500 men named Augustus.
(**) But there's 15,000 women named Augusta, 4,500 named Augustina, and 3,000 named Augustine.

There's no doubt other variants (especially if you count "Julie" or "Julius" as variants of "July"), but for simplicity and objectivity I only counted obvious variants that turned up in a string search for the primary form.

For comparison, the 10th most common male name was "Thomas", with an estimated male population of 2,050,000 and an estimated female population of 4,500, and the 10th most common female name was "Dorothy", with an estimated female population of 1,112,000 and an estimated male population of 0.

Comments:

[User Picture]
From:maniakes
Date:January 6th, 2012 10:19 pm (UTC)
(Link)
Caveats:
- "0" values are approximate, since unique or extremely rare names either wouldn't show up in the sample or would get elided in the Census Bureau's analysis process as being below the reporting threshold.

- The names file was a secondary collection of information from the PES, and its sample is skewed somewhat by the primary purpose of the PES: the PES was primarily intended to estimate the number of people missed by the full official census tally, and it deliberately over-sampled minority communities (especially blacks and hispanics) because those were the communities they were most concerned about being undercounted. Thus, the ethnic/cultural demographics of the raw PES sample aren't quite representative of the overall US population, and there was no attempt to correct the names files for the demographic skew.

- The data was collected in 1990, making it nearly 22 years out of date. Sadly, this is the most recent data set available.
(Reply) (Thread)
[User Picture]
From:mor_riogain
Date:January 9th, 2012 09:07 am (UTC)
(Link)
That's actually pretty damned cool!
(Reply) (Thread)
[User Picture]
From:squid314
Date:January 17th, 2012 05:28 pm (UTC)
(Link)
Very interesting indeed. I also feel I should add that a friend of a friend named her daughter Wednesday.
(Reply) (Thread)
[User Picture]
From:maniakes
Date:January 17th, 2012 06:02 pm (UTC)
(Link)
The lack of Wednesdays in the sample surprised me a bit. I would have expected at least a few, due to people naming daughters after Wednesday Friday Addams.
(Reply) (Parent) (Thread)
[User Picture]
From:maniakes
Date:January 17th, 2012 06:04 pm (UTC)
(Link)
Although now that I think to check, Pugsley, Morticia, Fester, Lurch, and even Gomez seem to all be absent from the sample.
(Reply) (Parent) (Thread)