Research Library
Most of Namesake's research sits behind the scenes — it quietly shapes the recommendations you see in the app. This page is for the rest of it: the full paper, the reviews it's getting, and every per-phase technical report the pipeline has produced.
We're publishing these as working drafts, not finished academic output. Some reports contain placeholder values that will firm up as later phases complete. They're here because the work itself is part of the story of the product, and anyone who wants to read it ought to be able to.
Looking for the plain-English version? Start with What Makes a Baby Name Go Viral? instead.
The paper
The formal multi-method write-up — Abadie synthetic controls, Hawkes processes, Bass diffusion, a Lieberson null model, and a Salganik-style predictability ceiling exercise, all applied to the same 145-year corpus. Draft status.
The full formal write-up — SSA birth registration, Google Trends, and cultural-event attribution combined through synthetic controls, Hawkes processes, Bass diffusion, and a Lieberson null model. Working draft.
Reviews & companion essays
A peer-reviewer's memo, an adversarial engineering critique of the pipeline itself, a product-and-strategy memo on what the findings do and don't buy us, and a long-form explainer written for a general reader. We published the unflattering ones too — that's the only honest way to do this.
A senior referee's review of the cultural-diffusion manuscript, written for a Sociological Science / ASR methods track. Candid, adversarial, constructive.
Adversarial PhD-level review of the orchestration and phases 3–7 — what's solid, what silently fails, and what to fix before publishing.
A long-form explainer for a general reader — inside a 145-year experiment in human imitation, and what the data does and doesn't tell you about the name on your shortlist.
What the cultural-diffusion pipeline actually buys us — which findings to ship, which to hold back, and which lines to cut from the marketing copy.
Per-phase technical reports
One report per analytical phase. These are the narrow, method-specific artifacts that feed the paper above.
A data analysis of 843 cultural events and their impact on baby naming across 145 years of SSA birth records
Do mega-hits produce fewer namesakes than moderately popular films? Hill-curve saturation modeling on the cultural-event panel.
Which kinds of names and events show the strongest cultural adoption effects? Systematic moderator analysis across the causal ATE panel.
Spatial autocorrelation of naming patterns across U.S. states, 1960–2024.
Does Google Trends search interest predict SSA birth rank changes? Granger tests on 4,185 names plus a pooled panel VAR.
How long a cultural shock keeps echoing through baby naming: branching ratios and half-lives from 6,328 Hawkes process fits.
Separating broadcast-driven from peer-driven adoption across 60,470 names via Bass diffusion (p, q) parameter fits.
Salganik-style forecasting exercise: how well can any model predict which names enter the SSA top 100?
Seven small-but-surprising tests benchmarked against the neutral-drift null — including the Blockbuster Paradox, phonetic saturation, and sibling-name avoidance.
Nested OLS attributing naming variance to event, name, phonetic, and generational-cycle features.
Data sources, analytical methods, and limitations for the Namesake research pipeline.
Full-scale Phase 5 analysis: 1.95M threshold rows across 46,412 names, 145 years of SSA data. Neutral drift + phonetic fashion null models with Wright-Fisher calibration.
Phase 6: phonetic neighborhood cross-correlation, clusters, Granger tests, and spillover magnitudes vs SSA panel.
Questions, collaboration, or citations? Reach us at hello@namesake.baby.