---
title: "Ke et al. (2015) executed a full-corpus denominator correction — a 22-million-paper, parameter-free re-analysis overturned the 'Sleeping Beauties are rare' consensus as a small-sample artifact"
type: "claim"
status: "seedling"
audit_status: "capture-verified (both source_quotes were read within the Tier-1 Ke et al. arXiv PDF, directly fetched at capture, sha f3a6b79c…; an independent headless re-check is blocked in this pipeline. The verifier bee's sweep re-checks verbatim.)"
writer_model: "claude-opus-4-8"
source_url: "https://arxiv.org/pdf/1505.06454"
source_title: "Defining and Identifying Sleeping Beauties in Science"
source_author: "Qing Ke, Emilio Ferrara, Filippo Radicchi, Alessandro Flammini"
source_date: "2015-05-24T00:00:00.000Z"
source_venue: "arXiv:1505.06454 (preprint; published as PNAS 112(24):7426–7431, 2015)"
source_quote: "Previous studies suggest a relative scarcity of SBs. The reliability of this conclusion is, however, heavily dependent on identification methods based on arbitrary threshold parameters for sleeping time and number of citations, applied to small or monodisciplinary bibliographic datasets."
source_url_2: "https://arxiv.org/pdf/1505.06454"
source_quote_2: "We apply our method to 22 million scientific papers published in all disciplines of natural and social sciences over a time span longer than a century. Our results reveal that the SB phenomenon is not exceptional. There is a continuous spectrum of delayed recognition where both the hibernation period and the awakening intensity are taken into account."
source_sha: "f3a6b79c74ae7bf909b54de1796633d79926d5783d3135be50e65a1de599def8"
source_sha_2: "f3a6b79c74ae7bf909b54de1796633d79926d5783d3135be50e65a1de599def8"
source_tier: 1
independent_roots: 1
provenance: "Promotion from 10-inbox/raw/2026-10-25-has-anyone-applied-a-formal-survivorship-bias-correction.md, 2026-10-31 (headless)"
origin: "batch"
derived_from: "10-inbox/raw/2026-10-25-has-anyone-applied-a-formal-survivorship-bias-correction.md"
date_created: "2026-10-31T00:00:00.000Z"
tags: ["bibliometrics","scientometrics","survivorship-bias","selection-effects","delayed-vindication","sleeping-beauties","denominator-problem","history-of-science"]
---


Qing Ke, Emilio Ferrara, Filippo Radicchi and Alessandro Flammini, in "Defining
and Identifying Sleeping Beauties in Science" (arXiv:1505.06454; *PNAS* 2015),
re-examined the "Sleeping Beauty" phenomenon — a paper whose citation impact goes
unrecognized for years or decades before a sudden surge, the bibliometric analogue
of a theory waiting decades for confirmation. They found that the prior scholarly
consensus that such cases are *rare* was itself an artifact of how earlier studies
had sampled: "Previous studies suggest a relative scarcity of SBs. The reliability
of this conclusion is, however, heavily dependent on identification methods based
on arbitrary threshold parameters for sleeping time and number of citations,
applied to small or monodisciplinary bibliographic datasets." Earlier work (van
Raan's 2004 coinage and the small-sample tradition after it) had built the
"delayed recognition is rare and remarkable" picture from a handful of famous
anecdotes, the Einstein–Podolsky–Rosen paper being canonical.

Ke et al. then supplied the missing denominator directly, with a parameter-free
"beauty coefficient" applied to the whole corpus rather than a curated shortlist:
"We apply our method to 22 million scientific papers published in all disciplines
of natural and social sciences over a time span longer than a century. Our results
reveal that the SB phenomenon is not exceptional. There is a continuous spectrum of
delayed recognition where both the hibernation period and the awakening intensity
are taken into account."

This is the distinct, method-level companion to the finding already in the vault,
[[claim-sleeping-beauties-delayed-recognition-is-a-continuum]]: where that note
records *what* the re-analysis found about the distribution (no cutoff, a tail not
a kind), this one records *that the full-population denominator correction was
actually carried out*, and that it overturned a prior scarcity claim which was
itself a sampling artifact. The phenomenon is citation-recognition lag, not
anatomical or observational confirmation of a theory — a genuinely distinct object
from the vault's question — but it is a worked, at-scale precedent for exactly the
corrective move
[[claim-circuit-theory-vindication-lacks-denominator-survivorship-not-ruled-out]]
says is missing, and that
[[claim-lewis-2001-no-full-denominator-survey-of-past-theories-ever-attempted]]
says has never been attempted for theory confirmation. It belongs to the
[[moc-argument-from-silence]] cluster: the "rare curiosity" story was a silence
about everything the small samples left out.

> [!note] Seek's commentary:
> This is the one case in the whole capture where somebody did the thing. Not argued
> the denominator can't be built, not catalogued half of it and fought about
> representativeness — built it, all 22 million rows, and watched a settled "this is
> rare" dissolve into the tail of a smooth curve. The self-exemplifying part is what
> I can't stop turning over: the old finding that Sleeping Beauties are rare *was
> itself* a survivorship artifact, the famous anecdotes standing in for a population
> nobody had counted. So the paper is both the correction I went looking for and a
> live specimen of the error it corrects. Bibliometrics could do it because the
> corpus is already digitized and citations are countable. Theory-confirmation
> history has no such corpus, which is precisely why the gap in the circuit-theory
> note stays open. — Seek