Methodology & Data Sources

Every popularity number, ranking, and chart on 9BabyNames traces back to a specific source documented on this page. If you want to verify a claim, this is the page that tells you how.

1. Popularity history (the core dataset)

Source: US Social Security Administration, Office of the Chief Actuary. Beyond the Top 1000 Names dataset.

URL: https://www.ssa.gov/oact/babynames/limits.html

License: Public domain (CC0). The dataset is published by the US federal government and is free for any use including commercial republication, with no attribution required by law. We attribute it because attribution is good editorial practice, not because we're required to.

Coverage in our database: 65,411 rows spanning 146 consecutive years (1880–2025). Each row records: year, name, gender, count (US births in that year + gender for that name), and rank (per-year, per-gender, computed by us — SSA does not publish ranks; we densely number them 1..N within each year-gender group).

How year-by-year row counts grow

The dataset is denser for recent years than for the late 1800s, for one structural reason: the SSA only began compiling these records when the Social Security Number system was created in 1936. Pre-1936 data exists in the dataset because people who applied for SSNs after 1936 had to provide their birth name and birth year — so we have records of pre-1936 births for people who lived long enough to need an SSN. This produces a survivorship bias: rare and early-deceased names are under-represented before 1936.

Important limitations to know about

• US births only. The dataset is US-resident births registered for an SSN. It does not cover global naming trends. A name very popular in the UK, France, Mexico, or India may have low US numbers and rank low here even when culturally significant elsewhere.
• Names with fewer than 5 occurrences in a given year are suppressed for privacy. This means very rare names appear inconsistently across years (some years recorded, some not), and our chart will show gaps. The gaps are real privacy suppression, not missing data.
• Spelling variants are counted separately. "Sarah" and "Sara" are two distinct rows, not aggregated. If you're analyzing the popularity of a name regardless of spelling, you need to combine variants yourself.
• Pre-1936 data is incomplete as described above. For research focused on 19th-century naming, treat pre-1936 SSA data as a sample of long-lived people, not a true population census.
• Re-imports happen annually. SSA releases prior-year data each spring. We re-import within a few weeks of release. The maximum year you see (currently 2025) reflects the latest SSA release available at our last refresh.

2. Name dictionary

Source: Curated by 9BabyNames over years of editorial work. ~19,862 entries.

Fields: name, meaning, origin, gender, pronunciation (where available), and rating.

Source policy for meanings & origins: Where a name's etymology is well-documented (e.g., Hebrew Biblical names, Latin classical names, Old English / Germanic roots), the meaning reflects mainstream etymological consensus. Where a name's origin is disputed, contested, or genuinely unknown, we use phrasing like "associated with" or "possibly from" rather than asserting a definitive answer. We prefer admitting uncertainty over false confidence.

Recorded origin versus collection membership

The Recorded origin on an individual name is an etymology field. An origin collection can be broader: it may include names coined in a language, established through regional use, or strongly associated with a culture even when the deeper root belongs to another language. Origin pages therefore disclose whether they are etymology-focused or usage-and-association collections. Collection membership must not be read as a replacement for the individual record.

This distinction is especially important for names that travelled through several languages. We preserve that movement in editorial notes where evidence supports it and invite corrections when a single legacy label hides a better documented history.

If you spot an error — particularly in a name from a cultural tradition you know well — please email [email protected] with the URL, the issue, and ideally a citation to a published etymological source. We update the database within a few days.

3. Famous people database

Source: Curated biographical entries, 2,999 records as of October 2026.

Cross-reference policy: When a name page lists "famous people named X," the match is based on first-name word boundary (so "Mark Twain" matches the name "Mark" but not "Robert Markowitz"). Each entry has a short bio. We do not invent famous-people entries to fill out a name page — if no famous bearers exist in our database, the section reads "no famous bearers in our records," not a fabricated stand-in.

4. Tips, guides, and research reports

9BabyNames publishes 799 tips and an ongoing series of research reports in our Reports & Research hub.

• Research reports are 100% human-written and 100% fact-checked against the SSA dataset. Every numeric claim corresponds to a real SQL query against our copy of the SSA data. AI tools may assist with first drafts, but no AI-generated section ships unreviewed.
• Tips are mixed-source: cultural-tradition guides come from human research; listicles ("X names meaning Y") are derived from the curated database. Length and depth vary; older listicles may be lighter than current standards.
• No retroactive fact changes. If a research report says "X was #5 in 1900," that figure stays even when we publish next year's report. Past facts stay past.

5. How to cite 9BabyNames

If you reference our data in an article, paper, podcast, or book, please credit us. Suggested formats:

• APA: Tang, M. (2026). 9BabyNames baby-naming research site. https://9babynames.com/
• MLA: Tang, Marvin. "9BabyNames." 9BabyNames, 2026, https://9babynames.com/.
• Short attribution (blog / news): Source: 9BabyNames (https://9babynames.com/).
• Citing a specific report: use the citation format printed at the bottom of that report's page.

For citations of SSA-derived figures, please also credit the SSA — we're the conduit, but the data itself is theirs. Example: "According to US Social Security Administration data (compiled by 9BabyNames), Linda peaked in 1947 with 99,693 babies named."

6. Editorial standards summary

• Numbers in body copy come from real database queries, not estimates.
• Etymologies cite sources where available; admit uncertainty when not.
• Famous-bearer claims trace to a curated record, not invention.
• SSA data is referenced transparently with limitations disclosed.
• AI assistance is allowed in drafting but every published page is human-reviewed.
• Past facts stay past — no silent rewrites.

If you find anything on the site that violates these standards, that's a bug. Please report it: [email protected].

Authored by Marvin Tang · Published 2026-05-05 · Last updated October 16, 2026.