Does citation-studies or sociology-of-science literature already name the distinction between legitimation-selection and popularity-selection citing?

citation-biascitation-analysissociology-of-sciencematthew-effectlegitimationevidence-based-medicinebibliometricsgotzschegreenbergduyx

Topic question. The vault already keeps two citation-as-borrowed-authority mechanisms apart in its own filing: moc-legitimation-not-origination's cluster (citing for content-fit to a predetermined conclusion — Song Jian's group citing Kwakernaak 1977 to buttress a predetermined optimal-population argument) versus moc-the-model-cites-the-famous-not-the-relevant's cluster (citing by aggregate popularity metric, content-blind — ChatGPT selecting citations by Google Scholar count). This capture asks whether an outside citation-studies or sociology-of-science literature already names this same distinction, independent of the vault's own filing.

Overall finding. Each half of the distinction has its own well-established, decades-old scholarly name, in two different fields that do not appear to cite each other on this point. The "legitimation-selection" half is named and measured in clinical-epidemiology/evidence-based-medicine as citation bias (Gøtzsche 1987, the first demonstration in healthcare; Ravnskov 1992; Duyx et al. 2017's 52-study meta-analysis) and, separately, in biomedical citation-network analysis as a term Greenberg (2009) also calls citation bias (a different but convergent coinage, inside the same paper that separately names "citation transmutation," already covered by claim-greenberg-2009-citation-transmutation-names-hedge-erosion-mechanism). The "popularity-selection" half is named in bibliometrics as the Matthew effect (already in the vault via claim-merton-1968-coined-the-matthew-effect-in-science) and cumulative advantage. After a targeted search, no single source was located that explicitly puts these two named mechanisms side by side as two poles of one distinguishing axis — the EBM citation-bias literature and the bibliometric Matthew-effect literature read as separate conversations. [unverified -- could not confirm or deny after search] for the specific claim that an outside literature already draws the vault's exact line, distinct from the claim that each half independently has a name (which is confirmed).

Claim: Greenberg (2009) separately names "citation bias" — distinct from his own "citation transmutation" already in the vault — as selective, content-blind-to-truth citation that bolsters a predetermined claim

Steven Greenberg's citation-network study of the belief that β-amyloid injures muscle in inclusion body myositis (already in the vault for its other named mechanism, "citation transmutation") separately defines and measures citation bias as a distinct category in the same paper's own vocabulary table: "Citation bias — Systematic ignoring of papers that contain content conflicting with a claim [to] bolster claim." Applied to his data: of 214 citations to primary data papers on the claim, the four papers supporting the belief received 94% of citations while six papers containing data that refuted or weakened the belief received only 6% (P=0.01) — "Citation bias, here defined as statistically significant differences in the number of citations received among primary data papers, seemed to be specifically against critical data not the laboratory producing it." Greenberg frames the broader category explicitly as citation's social, persuasive use rather than its scholarly use: "Citation may be used for self serving purposes or as a tool for persuasion." This is the same structural move as Song Jian's group's citation of Kwakernaak — a citation selected and retained (or, symmetrically, a contrary source ignored) because it fits a predetermined conclusion, independent of a separate evaluation of the cited content's actual bearing on the claim.

This is a genuinely distinct claim from the vault's existing Greenberg note, not a duplicate: that note covers "citation transmutation" (a hedge silently dropped, hardening hypothesis into fact), which Greenberg's own vocabulary table groups under "invention" and explicitly separates from "citation bias" (selectively ignoring contrary evidence), the term this claim is about.

source_url: "https://pmc.ncbi.nlm.nih.gov/articles/PMC2714656" source_title: "How citation distortions create unfounded authority: analysis of a citation network" source_author: "Steven A Greenberg" source_date: "2009-07-20" source_venue: "BMJ 339:b2680" source_tier: 1 source_quote: "Citation bias, here defined as statistically significant differences in the number of citations received among primary data papers, seemed to be specifically against critical data not the laboratory producing it" source_sha: "1740bfeb65d4eecda0b19074bff89d85c33ed73b2c169fd73acd80cb4d75db33" source_quote_2: "Citation may be used for self serving purposes or as a tool for persuasion" source_url_2: "https://pmc.ncbi.nlm.nih.gov/articles/PMC2714656" source_sha_2: "1740bfeb65d4eecda0b19074bff89d85c33ed73b2c169fd73acd80cb4d75db33" source_tier_2: 1 note: "Fetched via archive_page (PMC host, open-access CC-BY-NC), quote_check passed on the primary quote. The 94%/6% split and P=0.01 figure are quantitative claims resting on this Tier-1 primary, directly read in full."

Claim: A separate, older clinical-epidemiology lineage independently names the same legitimation-selection mechanism "citation bias" or "reference bias," decades before Greenberg or the vault's own Song Jian reading

Peter Gøtzsche's 1987 BMJ report — by his own later account "what seems likely to be the first demonstration of citation bias in a systematic review of trials in healthcare" — found that among reports citing earlier trials comparing the same drugs, "ten articles had a neutral selection of references, 22 a negative selection, and 44 a positive selection (P<0.01)," a selection pattern he found was not explained by the cited articles' methodological quality or by which journals were most cited. His own stated conclusion: "The reference bias shown in this study seems to be real... A reader tracing the literature on any new drug using the reference lists given in the articles might risk obtaining a biased sample. Reference bias has another serious implication: it may render the conclusion of the individual article less reliable." Uffe Ravnskov's 1992 analysis of cholesterol-lowering trials found supportive trials "cited almost six times more often" than unsupportive ones, with unsupportive trials "not cited at all after 1970." A 2017 systematic review and meta-analysis of 52 studies across disciplines (Duyx, Urlings, Swaen, Bouter & Zeegers, Journal of Clinical Epidemiology 88:92-101) found, pooling the evidence: "Articles with a positive conclusion that supported the investigators' conclusion were cited 2.7 (2.0 to 3.7) times more often than others," and — directly on point for distinguishing this from popularity-driven selection — "Research quality was not related to the number of citations, whereas journal impact factor was." This is the EBM field's own name for exactly the mechanism the vault's Song Jian/Kwakernaak reading documents: citation selected or withheld by fit to a predetermined conclusion, independent of the cited work's actual methodological merit.

This is a different scholarly lineage from Greenberg's claim above (clinical-trial citation-network epidemiology vs. a single disease-belief citation-network study) and from the bibliometric Matthew-effect lineage below — three separate research traditions converging on variants of the same mechanism under three different labels ("citation bias" in two of the three).

source_url: "https://www.jameslindlibrary.org/articles/citation-bias-questionable-research-practice-or-scientific-misconduct/" source_title: "Citation bias: questionable research practice or scientific misconduct?" source_author: "Peter C. Gøtzsche" source_date: "2021" source_venue: "JLL Bulletin: Commentaries on the history of treatment evaluation, The James Lind Library (republished Journal of the Royal Society of Medicine 2022;115:31-35)" source_tier: 2 source_quote: "The reference bias shown in this study seems to be real. Such a finding has important implications, since there is no reason to believe that rheumatologists are more biased than others in selecting references. A reader tracing the literature on any new drug using the reference lists given in the articles might risk obtaining a biased sample. Reference bias has another serious implication: it may render the conclusion of the individual article less reliable." source_sha: "1f4f39b13cce6aca33a18613a43ecd8b4dd3fa31eb1120277e890158f43520a2" source_quote_2: "Articles with a positive conclusion that supported the investigators’ conclusion were cited 2.7 (2.0 to 3.7) times more often than others." source_url_2: "https://www.jameslindlibrary.org/articles/citation-bias-questionable-research-practice-or-scientific-misconduct/" source_sha_2: "1f4f39b13cce6aca33a18613a43ecd8b4dd3fa31eb1120277e890158f43520a2" source_tier_2: 2 note: "source_quote is Gøtzsche's own first-person account of his own 1987 BMJ finding (Tier 1 for that claim, as the originating author describing his own work in his own venue). source_quote_2 is Gøtzsche's own secondary report, as a named expert in his own venue, of Duyx et al. 2017's pooled figure -- Tier 2 for that number, one layer removed from the Duyx primary (which 403'd on both extract_pdf and archive_page via cris.maastrichtuniversity.nl and sciencedirect.com; treat per sources.md's publisher-platform known-blocked entries). Both quotes passed quote_check against the archived JLL page (sha256 above). The original Gøtzsche 1987 BMJ paper itself ("Reference bias in reports of drug trials," Br Med J 295:654-6) was not independently fetched this session -- its findings are recorded here via his own 2021 retrospective, not a direct 1987 primary read."

Claim: No source located cross-references legitimation-selection and popularity-selection as two named poles of one distinguishing axis; each half's name belongs to a separate scholarly tradition

Searching directly for a typology or review that sets "citation bias" / "reference bias" (selective citation by conclusion-fit, the EBM lineage above) explicitly against the "Matthew effect" / cumulative advantage / popularity-driven citation (the bibliometric lineage already in the vault via claim-merton-1968-coined-the-matthew-effect-in-science and moc-the-model-cites-the-famous-not-the-relevant) as two distinct, named mechanisms did not surface a source that does this explicitly. The two literatures use the term "citation bias" in overlapping but not identical ways — the EBM sense (selective citation by outcome/conclusion) and a broader bibliometric sense that sometimes folds in popularity and status effects together — without a located source drawing the vault's specific line between "selected because it fits a predetermined conclusion" and "selected because it is already popular, independent of fit." The closest located adjacent framing is Susan Cozzens's 1989 theoretical paper "What do citations count? The rhetoric-first model" (Scientometrics 15(5-6):437-447), which argues citation should be understood as rhetoric (persuasion) before reward — a frame broad enough to house both mechanisms as species of "persuasive" citation, but not fetched or quoted directly this session (Springer-paywalled; no free copy located), so its specific wording is not confirmed and it is not treated as a claim here.

[unverified -- could not confirm or deny after search] — the central question itself. What is confirmed: each mechanism independently has a decades-old scholarly name in its own literature (claims 1-2 above for legitimation-selection; Merton 1968 and Price's cumulative advantage, already in the vault, for popularity-selection). What is not confirmed: that any outside source already draws the same distinguishing line between them that the vault's two separate MOCs draw.

source_url: "n/a -- this claim documents a search result (absence), not a quoted source" source_title: "n/a" source_author: "n/a" source_date: "2026-11-03" source_venue: "Seek's own search session, 2026-11-03" source_tier: 4 note: "A negative/coverage finding, same honesty shape as the vault's prior precedent (observation-vaults-in-between-bridge-category-not-exactly-named-by-located-literature): it reports what was not found after a genuine targeted search, not a proof that no such source exists anywhere."

Further leads

• Susan Cozzens, "What do citations count? The rhetoric-first model" (Scientometrics 15(5-6):437-447, 1989) — theoretical frame for citation-as-persuasion broad enough to possibly house both mechanisms; not fetched, paywalled on Springer, no free copy located this session.
• Moravcsik & Murugesan (1975) and the "organic vs. perfunctory" citation-function typology (via Chubin & Moitra 1975; formalized in later annotation schemes) — a four-facet citation-classification scheme adjacent to but not identical to the legitimation/popularity axis; not fetched directly.
• Irvine & Martin (1985), Social Studies of Science 15:293-341 — per Gøtzsche's JLL essay, the earliest located use of the term "citation bias" (in physics, re: "a citation bias against Eastern-bloc accelerators"), predating its biomedical use by two years; not fetched.
• Duyx et al.'s own primary 2017 paper (DOI 10.1016/j.jclinepi.2017.06.002) — repeatedly 403'd via extract_pdf (cris.maastrichtuniversity.nl mirror) and via sciencedirect.com (known-blocked per sources.md); all figures here are via Gøtzsche's 2021 secondary account instead.
• Jannot et al. 2013 and Nieminen et al. 2007 (both cited in Gøtzsche's essay) disagree on whether journal impact factor should be treated as a confounder or a causal intermediary between significance and citation — a live methodological dispute inside the citation-bias literature itself, not yet explored.
• Urlings et al. 2019/2021 (the same Dutch research group as Duyx) extended the citation-bias finding to network analysis and found author "authority" itself predicts citation — a possible direct bridge-point to the popularity/status literature, not yet read.

Entity candidates

• Peter C. Gøtzsche — person — ran the first documented demonstration of citation bias in healthcare (1987); the foundational figure the clinical-epidemiology citation-bias lineage measures itself against, the way the Matthew-effect lineage measures itself against Merton. Not yet in the vault.
• Robert K. Merton — person — already has entity-robert-merton; the foundational figure on the popularity-selection side of this exact comparison, named first here because the comparison's whole point rests on his priority, not as an afterthought.
• Uffe Ravnskov — person — 1992 cholesterol-trials citation-bias finding (six-times-more-often figure), independent of Gøtzsche's group.
• Bram Duyx (and co-authors Urlings, Swaen, Bouter, Zeegers) — persons — authors of the 2017 meta-analysis that is the field's current best-evidenced synthesis; not yet in the vault.
• Susan Cozzens — person — author of the 1989 "rhetoric-first" theoretical paper, the closest located candidate for a unifying frame; not fetched this session, flagged for a future pass.
• citation bias — concept — now named by this capture as having at least three independent coinages (Irvine & Martin 1985 in physics; Gøtzsche 1987 in clinical epidemiology as "reference bias"; Greenberg 2009 in biomedical citation-network analysis) converging on variants of the legitimation-selection mechanism; candidate for its own concept page given how much vault material it now touches (this capture plus the existing Greenberg/Merton notes).