SSA applicant totals changed the order of female and male top-ten name shares for 2015 and 2016 in babynames 1.0.1
SSA omits name counts below five. We tested whether the total used in a name share changes comparisons between female and male records. We used 31 birth years, from 1987 to 2017, in the pinned babynames 1.0.1 data. The analysis includes 922,600 name, sex, and year rows. Female top-ten shares were higher in 2015 and 2016 when the total included only published name records. The order changed when the total included all eligible applicants. In 2015, the female minus male difference changed from +0.1230 to -0.0747 percentage points. Other order changes occurred for the first name in 2005 and the first five names in 2009. There were no order changes for lists of 20, 50, 100, or 1000 names. We also calculated exact bounds for omitted counts of one to four per name. Female name-match rates were lower in every tested year under all allowed allocations. Female Shannon and inverse-concentration diversity were also higher under all allowed allocations. Bounds for the number of distinct names overlapped in every year. This is an exploratory audit of one data version. It does not identify why parents chose names. The package includes source files, hashes, code, figures, tables, and repeat instructions.
name choice · cultural change · SSA · missing data · diversity · denominator · identification bounds
References
5- usesdoi:10.32614/CRAN.package.babynames
We use the pinned 1.0.1 name counts, applicant totals, and source definitions.
- doi:10.1177/1948550609349515
This study used common-name shares to study US naming trends through 2007.
- doi:10.1111/jasp.12409
This study extended common-name trend analysis through 2015.
- doi:10.1179/175622710X12590782368107
This study compared name diversity and discussed limits of short common-name lists.
- doi:10.48550/arXiv.2103.14178
This work studied name frequency change and bounded missing counts for a different inference problem.
Cited by
5 publicationsPublic catalog entries on Sciscape.
- usesWith exact totals, SSA tables allow opposite changes in annual name turnover between 1980–1989 and 2008–2017
We use its independently prepared name-count export to check all 922,600 shared source rows. Its within-year diversity audit motivates the separate turnover question.
- usesTen prior birth years gave lower Brier loss than all prior years since 1950 in listed US name counts for 2000 to 2017
We reuse its retained, pinned babynames 1.0.1 source files and provenance. We use new analysis code.
- With exact totals, the first count threshold that leaves name entropy order unresolved ranges from 49 to 299 across paired SSA tables for 1980–2017
[Legacy relation: reproduces] The independent calculation matches ten diversity fields in all 62 shared year-sex cells. The new study adds artificial release thresholds, exact boundary checks, seven earlier years and endpoint changes.
- usesAn 80-year history input reduced squared log-share error by 0.52% in ten-year ridge forecasts of US names for 2007–2017
We reuse its retained original babynames 1.0.1 files and matching applicant totals, with source checksums. Its denominator audit informs our data handling.
- usesWith exact totals, SSA tables allow opposite changes in annual name turnover between 1980–1989 and 2008–2017
We use its independently prepared name-count export to check all 922,600 shared source rows. Its within-year diversity audit motivates the separate turnover question.