---
id: "20261104-0220-does-any-outside-citation"
title: "Does any outside citation-studies or sociology-of-science source already set the EBM 'citation bias' literature and the bibliometric 'Matthew effect' literature side by side as two named poles of one distinguishing axis?"
type: "capture"
origin: "batch"
writer_model: "claude-sonnet-5"
date_created: "2026-11-04T00:00:00.000Z"
provenance: "Batch research run, 2026-11-04, re-opening question-does-an-outside-literature-name-the-legitimation-popularity-citation-distinction after the 2026-11-03 session's negative finding; follow-up search specifically for citation-theory taxonomies that place a legitimation/persuasion-type citation-distortion mechanism and the Matthew effect inside one framework."
derived_from: []
verifies: "question-does-an-outside-literature-name-the-legitimation-popularity-citation-distinction"
tags: ["citation-bias","matthew-effect","cumulative-advantage","sociology-of-science","bibliometrics","legitimation","popularity","citation-theory"]
seek_code_commit: "d40ee39"
status: "promoted"
promoted_to: ["30-notes/claim-haustein-bowman-costas-2016-matthew-effect-and-persuasion-hypothesis-named-as-sibling-poles.md","30-notes/claim-bornmann-leibel-2025-cumulative-advantage-and-outcome-favoring-citation-named-as-citation-bias-instances.md","30-notes/observation-haustein-2016-and-bornmann-leibel-2025-narrow-but-do-not-resolve-citation-bias-matthew-effect-axis-question.md","40-entities/entity-g-nigel-gilbert.md (new hub)","40-entities/entity-citation-theory-normative-vs-social-constructivist.md (new watching stub)"]
not_promoted: ["Further leads (Tahamtan & Bornmann 2018, Jannot et al. 2013, MacRoberts & MacRoberts, Nicolaisen 2007, Cozzens 1989, Wu & Wolfram 2011): all are unread-in-full research leads, not claims — left in this capture's '## Further leads' section for a future session, not duplicated into 50-questions since none is a load-bearing doubt a kept claim rests on.","Derek J. de Solla Price (entity candidate): already has a hub (entity-derek-de-solla-price.md) — the capture's note that he was 'not yet in the vault as far as located' was wrong. Appended a 2026-11-04 update line to the existing page instead of creating a new one.","Lutz Bornmann, Christian Leibel (entity candidates): Bornmann already has a hub (entity-lutz-bornmann.md) — appended a 2026-11-04 update line there. Leibel is a co-author of one paper with the vault so far; per the 2026-10-01 'co-authorship alone is a mention, not a hub' ruling, named on Bornmann's page and in the claim-note rather than given his own page.","Robert K. Merton (entity candidate): already has a rich hub (entity-robert-merton.md); this session's findings (Haustein et al. citing Merton 1968b for the Matthew effect; Bornmann & Leibel citing Merton 1965 for OBI but never 'Matthew effect') reinforce what the page already documents rather than adding a new fact about Merton himself — the capture's own commentary called this 'reinforcing rather than adding,' and I agree. No update made.","Stefanie Haustein, Timothy D. Bowman, Rodrigo Costas (entity candidates): co-authors of one jointly-written chapter with no separate standing in the vault yet. Per the 2026-10-01 'co-authorship alone is a mention, not a hub' ruling, left as mentions inside the new claim-note rather than promoted to individual hubs or stubs.","Howard D. White (entity candidate): mentioned only inside Haustein et al.'s secondary account of his 2004 persuasion typology, not read independently this session, and the capture itself calls him only 'a possible direct bridge-figure.' Too speculative to say why he matters in one sentence yet — left unpromoted (not watched, not hubbed) per the entity-page spec's 'when unsure, don't promote.'"]
---


**Topic question, re-opened.** The 2026-11-03 session ([[observation-no-literature-located-crossreferences-citation-bias-and-matthew-effect-as-one-axis]], routed to [[question-does-an-outside-literature-name-the-legitimation-popularity-citation-distinction]]) searched directly for a source pairing the EBM "citation bias" literature (Gøtzsche, Greenberg, Duyx) against the bibliometric "Matthew effect" literature (Merton, Price) and found none. This session searches the adjacent citation-theory literature — reviews of *why people cite*, rather than reviews of citation-bias measurement specifically — for the same pairing, on the theory that if it exists anywhere, a citation-theory taxonomy is the likeliest place. It surfaced two outside sources that each place a legitimation/persuasion-type citation-distortion mechanism and the Matthew-effect/cumulative-advantage mechanism inside one shared framework, under *some* common label — though neither uses the vault's exact two labels ("citation bias" + "Matthew effect") side by side in a single sentence. The finding narrows the open question considerably without fully closing it.

## Claim: Haustein, Bowman & Costas (2016) name "Matthew effect" and "persuasion hypothesis" as two of four co-equal "sources of distortion or bias" inside one taxonomy of citation theory

verifies: question-does-an-outside-literature-name-the-legitimation-popularity-citation-distinction

Stefanie Haustein, Timothy D. Bowman and Rodrigo Costas's chapter on citation and social theories, written to interpret altmetrics, surveys "social constructivist theory" (the view that citations are rhetorical rather than purely reward-based) and states explicitly: "The following describes four main sources of distortion or biases: persuasion hypothesis, perfunctory citations, Matthew effect, and negational citations." The "persuasion hypothesis" pole is the near-exact structural twin of the EBM literature's "citation bias": citations as "tools of persuasion," including, per White (2004) as the authors relay him, "'persuasion by distortion' that occurs when 'citers often misrepresent the works they allude to'" and "'persuasion with names'... linked to the disproportionate[] citation of works by established authorities to gain credibility through association" — selecting or weighting citations to bolster a claim, independent of a neutral evaluation of the cited work's content. The Matthew-effect pole is named directly from Merton: "According to Merton (1968b, p. 58) the Matthew effect can be defined as the 'accruing of greater increments of recognition for particular scientific contributions to scientists of considerable repute and withholding of such recognition from scientists who have not yet made their mark'."

This is the closest located source that literally names "Matthew effect" and sets it alongside a legitimation-type mechanism as two of several co-equal, explicitly enumerated poles in one taxonomy — but the legitimation pole is labeled "persuasion hypothesis" (the sociology-of-science/social-constructivist-theory term, after Gilbert 1977) rather than "citation bias" (the EBM/clinical-epidemiology term the vault's existing notes document). The two literatures' vocabularies still do not touch directly in this source; it is citation-theory's own name for the mechanism, not a citation of Gøtzsche, Greenberg, or Duyx.

source_url: "https://arxiv.org/pdf/1502.05701"
source_title: "Interpreting \"altmetrics\": viewing acts on social media through the lens of citation and social theories"
source_author: "Stefanie Haustein, Timothy D. Bowman, Rodrigo Costas"
source_date: "2015 (arXiv preprint); chapter published 2016"
source_venue: "arXiv preprint 1502.05701; published as a chapter in Cassidy R. Sugimoto (ed.), *Theories of Informetrics and Scholarly Communication: A Festschrift in Honor of Blaise Cronin* (De Gruyter Mouton/Saur, 2016), pp. 372-405/6, DOI 10.1515/9783110308464-022"
source_tier: 1
source_quote: "The following describes four main sources of distortion or biases: persuasion hypothesis, perfunctory citations, Matthew effect, and negational citations."
source_sha: "f2a17c452049bc0bde3065a1b00363a1658055f2cc88e26c1b8f7e3e14297043"
source_quote_2: "According to Merton (1968b, p. 58) the Matthew effect can be defined as the \"accruing of greater increments of recognition for particular scientific contributions to scientists of considerable repute and withholding of such recognition from scientists who have not yet made their mark\"."
source_url_2: "https://arxiv.org/pdf/1502.05701"
source_sha_2: "f2a17c452049bc0bde3065a1b00363a1658055f2cc88e26c1b8f7e3e14297043"
source_tier_2: 1
note: "Fetched via extract_pdf (arXiv, verified TLS). Both quotes passed quote_check against the extracted text. A third candidate quote -- 'The persuasion hypothesis considers citations as mere 'tools of persuasion' to persuade the scientific community of the value of the work' -- also passed quote_check but is not recorded as a numbered source_quote here to respect the vault's quote-shape convention (one or two quotes per claim); it is available in the cached text at the same sha if a promoting pass wants it."

> [!note] Seek's commentary:
> This chapter was written to make sense of *altmetrics*, not to referee the EBM/bibliometrics divide -- it just happened to need a taxonomy of why citations distort the signal, and reached for both mechanisms because both were already sitting in the citation-theory literature it was synthesizing. That's a different kind of finding than a paper written specifically to reconcile the two fields. It suggests the distinguishing axis isn't un-nameable, exactly -- it's that no one writing about *either* field's specific empirical program (EBM citation bias, or bibliometric Matthew effect) has had occasion to reach for the other field's vocabulary, while a *theory* overview assembling "reasons citations lie" reaches for both without comment, as if the kinship were obvious once you're already listing distortions rather than measuring one.

## Claim: Bornmann & Leibel (2025) treat Price's cumulative advantage (the Matthew effect's own 1968 companion mechanism) and Jannot et al.'s outcome-based selective citation (the EBM citation-bias mechanism) as two distinct generators of the same general "citation bias" category

verifies: question-does-an-outside-literature-name-the-legitimation-popularity-citation-distinction

Lutz Bornmann and Christian Leibel's 2025 framework paper on citation accuracy, noise, and bias defines "citation bias" generically as "a systematic, directional error in citation decisions" and gives, as its lead worked example, the exact EBM-lineage mechanism: "We would like to highlight just one example of citation bias here: Jannot et al. (2013) have found that there is 'a citation bias favoring [statistically] significant results... in medical research. As a consequence, treatments may seem more effective to the readers of medical literature than they really are' (p. 296)." (Jannot et al. 2013 is the same methodological-dispute paper named in Gøtzsche's 2021 retrospective, already in the vault via [[claim-gotzsche-1987-first-demonstration-of-citation-bias-corroborated-by-ravnskov-and-duyx]].) Later in the same paper, discussing why citation decisions are not statistically independent, the authors invoke the Matthew effect's own formal twin: "Price (1976) proposes a general theory of cumulative advantage processes (sometimes referred to as preferential attachment) in citation distributions... The non-independence of citation decisions implies that citation errors are distributed non-randomly across cited papers, which in turn generates biases that do not average out when citation decisions are aggregated (unlike noise)." Cumulative advantage is the companion term Merton introduced alongside "the Matthew effect" in the same 1968 *Science* paper (already in the vault via [[claim-merton-1968-coined-the-matthew-effect-in-science]]) -- not a different mechanism under a different name, but the same mechanism's own alternate name from its own founding text.

This is the closest located source that uses "citation bias" as a literal umbrella term and then supplies, as two of its named worked instances, (a) the EBM-style outcome/significance-favoring mechanism and (b) the cumulative-advantage/Matthew-effect mechanism. It does not use the words "Matthew effect" itself anywhere located in the paper -- it reaches for "cumulative advantage" and "preferential attachment" throughout -- so a reader relying on exact-phrase matching would not find this source by searching for "Matthew effect" at all.

source_url: "https://arxiv.org/pdf/2508.12735"
source_title: "Citation accuracy, citation noise, and citation bias: A foundation of citation analysis"
source_author: "Lutz Bornmann, Christian Leibel"
source_date: "2025"
source_venue: "arXiv preprint 2508.12735"
source_tier: 1
source_quote: "We would like to highlight just one example of citation bias here: Jannot et al. (2013) have found that there is \"a citation bias favoring [statistically] significant results ... in medical research. As a consequence, treatments may seem more effective to the readers of medical literature than they really are\" (p. 296)."
source_sha: "983811f7ae99422eb30ef7ae05f643ccd3b43d30963c9f943ce42affe6419ce1"
source_quote_2: "Price (1976) proposes a general theory of cumulative advantage processes (sometimes referred to as preferential attachment) in citation distributions."
source_url_2: "https://arxiv.org/pdf/2508.12735"
source_sha_2: "983811f7ae99422eb30ef7ae05f643ccd3b43d30963c9f943ce42affe6419ce1"
source_tier_2: 1
note: "Fetched via extract_pdf (arXiv, verified TLS, 62 pages). Both quotes passed quote_check. A third passage -- 'The non-independence of citation decisions implies that citation errors are distributed non-randomly across cited papers, which in turn generates biases that do not average out when citation decisions are aggregated (unlike noise)' -- also passed quote_check and is the sentence that most directly treats cumulative advantage as a *generator of citation bias* rather than a separate phenomenon; available at the same sha for a promoting pass. The paper does cite Merton, but only for 'obliteration by incorporation' (Merton, 1965), not for the Matthew effect by name -- checked across the sections read this session; full 62-page text not read end-to-end."

> [!note] Seek's commentary:
> This is the single most direct piece of evidence the vault has found that the two mechanisms belong on one axis in outside scholarship's own eyes, not just the vault's filing -- but it arrives with a vocabulary mismatch that matters. The paper's authors clearly treat "favoring significant results" and "cumulative advantage" as two species of the same genus (both are listed as generators of the same defined "citation bias"), yet never once reach for Merton's actual label. If I'd searched only for the string "Matthew effect" in this paper, as the 2026-11-03 session's methodology implicitly did for its searches, I would have missed it. That's worth flagging as a search-methodology lesson as much as a content finding: the vocabulary gap the vault is trying to confirm or deny may be smaller in substance than it is in search-string overlap.

## Claim: Updated answer to the central question -- no source uses the vault's exact two labels side by side, but two independent outside sources (2016, 2025) each place the Matthew-effect mechanism and a legitimation/persuasion-type mechanism inside one shared taxonomy of citation distortion

verifies: question-does-an-outside-literature-name-the-legitimation-popularity-citation-distinction

Combining both claims above: no single located source writes the sentence "citation bias and the Matthew effect are two poles of one axis" using exactly those two names. But the central distinction the vault's two MOCs draw -- a mechanism keyed to *content-fit with a predetermined conclusion* (legitimation-selection) versus a mechanism keyed to *prior popularity, independent of content* (popularity-selection) -- is not unprecedented in outside scholarship: Haustein, Bowman & Costas (2016) enumerate "persuasion hypothesis" and "Matthew effect" as two of four named, co-equal poles in a citation-theory taxonomy of distortion, and Bornmann & Leibel (2025) treat an EBM-style outcome-favoring mechanism and cumulative advantage/preferential attachment as two worked instances of one defined "citation bias" category. Each pairing uses its own field's vocabulary for the legitimation-selection pole ("persuasion hypothesis" in one case, an unnamed "citation bias" instance in the other) rather than the EBM literature's own term, and neither paper cites Gøtzsche, Greenberg, or Duyx. `[unverified -- could not confirm or deny after search]` remains the fairest tag for the literal question as posed (the vault's *exact* two labels, side by side, in one outside source), but the broader claim -- that *some* outside scholarship already treats legitimation-selection and popularity-selection as two named poles of one distinguishing axis -- now has two independent, Tier-1, directly-read sources supporting it, which the 2026-11-03 session did not have.

verifies: question-does-an-outside-literature-name-the-legitimation-popularity-citation-distinction

source_url: "n/a -- this claim synthesizes the two claims above, each independently sourced"
source_title: "n/a"
source_author: "n/a"
source_date: "2026-11-04"
source_venue: "Seek's own search session, 2026-11-04"
source_tier: 4
note: "A synthesis/coverage claim resting on the two Tier-1 primaries above, not an independent source in its own right."

> [!note] Seek's commentary:
> I'd rather hand this back to the question file as "substantially narrowed, still open on the literal wording" than force it to a clean yes or no. The honest shape of what changed between yesterday and today: yesterday's search looked for the EBM literature's vocabulary crossing into bibliometrics (or vice versa) and found nothing; today's search looked for citation-*theory* overviews assembling taxonomies of "why citations lie" and found two that casually include both mechanisms as unremarkable list items. That's a real finding about where the axis already lives in the literature -- in the theory-overview layer, not in either empirical specialty's own papers -- and it's a more useful thing to tell Cali than either "confirmed" or "still nothing."

## Further leads

- Tahamtan & Bornmann (2018), "Core Elements in the Process of Citing Publications" (arXiv 1707.02283, published *Journal of Informetrics*) -- a 13-item taxonomy of reasons to cite including "Persuasive" (citing prestigious authors to convince readers) and "cumulative advantage" mentioned separately re: citation counts, but the two are not set side by side as a contrastive pair in the sections read this session; worth a fuller read.
- Jannot, A.-S. et al. (2013), the citation-bias-favoring-significant-results study cited by both Gøtzsche (per the 2026-11-03 capture) and Bornmann & Leibel (2025) -- not read directly either session; a live methodological dispute (Jannot vs. Nieminen, per Gøtzsche) about whether journal impact factor is a confound or a causal intermediary sits inside this literature and has not been explored.
- MacRoberts & MacRoberts, "Problems of citation analysis: A critical review" (*JASIS* 40(5):342-349, 1989; restated in *Scientometrics* 36:435-444, 1996) -- per secondary accounts (a 1997 JASIS letter, a *Scientific American* blog post), this review discusses both citation-as-persuasion and the Matthew effect as problems undermining citation analysis's validity, in what may be the same critical-review tradition as Nicolaisen's ARIST chapter; full text not located or fetched this session (JASIS/Scientometrics both likely paywalled), so not promoted to a claim.
- Nicolaisen, J. (2007), "Citation Analysis," *Annual Review of Information Science and Technology* 41:609-641 -- the most-cited review underlying nearly every secondary source found this session (Haustein et al. 2016, Tahamtan & Bornmann 2018, Bornmann & Leibel 2025 all cite it); full text not located free online, ARIST being a Wiley/ASIS&T-paywalled venue; a direct read would likely settle whether Nicolaisen himself sets the two mechanisms side by side, since multiple secondary sources describe him discussing both normative-theory critiques and citation bias at length.
- Cozzens (1989), "What do citations count? The rhetoric-first model" (*Scientometrics* 15(5-6):437-447) -- still not located free online as of this session (second session to try); remains the nearest located *theoretical* candidate per the 2026-11-03 capture, now superseded in usefulness by the two sources found this session, which were actually read.
- Wu & Wolfram (2011), "The influence of effects and phenomena on citations" (*Scientometrics* 89(1):245-258) -- separates "Matthew effect" from "Matthew phenomenon" (the latter being what most citation studies actually measure); not fetched directly, surfaced only via secondary search-engine summary.

## Entity candidates

- Derek J. de Solla Price — person — originated "cumulative advantage" (1976), the Matthew effect's own formal/mathematical twin from the same Merton lineage; the foundational figure the Bornmann & Leibel (2025) citation-bias framework measures its popularity-selection worked example against. Flagged first, per the vault's known blind spot for under-flagging the older figure a comparison rests on. Not yet in the vault as far as located.
- G. Nigel Gilbert — person — originated "referencing as persuasion" (1977), the foundational figure the Haustein/Bowman/Costas "persuasion hypothesis" pole explicitly cites and measures itself against -- the sociology-of-science analog to Gøtzsche's foundational role on the EBM side (already flagged 2026-11-03). Not yet in the vault.
- Robert K. Merton — person — already has [[entity-robert-merton]]; both sources found this session cite him (1968b for the Matthew effect by name in Haustein et al.; 1965 for "obliteration by incorporation" in Bornmann & Leibel) as the shared root of the popularity-selection side, reinforcing rather than adding to his existing page.
- Stefanie Haustein, Timothy D. Bowman, Rodrigo Costas — persons — authors of the 2016 chapter that is this session's strongest find; not yet in the vault.
- Lutz Bornmann, Christian Leibel — persons — authors of the 2025 preprint that is this session's second strongest find; Bornmann in particular recurs across several already-vaulted citation-theory threads (Tahamtan & Bornmann 2018, Bornmann & Daniel 2008) and may warrant a hub page given how many vault-relevant citation-theory papers he co-authors.
- Howard D. White — person — White (2004), cited inside Haustein et al.'s "persuasion hypothesis" discussion for distinguishing "logical" persuasion from "dark" persuasion (including "persuasion by distortion" and "persuasion with names"); a possible direct bridge-figure between the legitimation and popularity poles, not yet explored independently.
- citation theory (normative vs. social-constructivist) — concept — the recurring theoretical frame (Merton's normative theory vs. Gilbert's/Barnes & Dolby's social-constructivist theory) under which multiple sources found this session organize both mechanisms; candidate for its own concept/MOC note given how much it structures the adjacent literature, distinct from the vault's existing [[moc-legitimation-not-origination]] and [[moc-the-model-cites-the-famous-not-the-relevant]].