---
title: "Garfield grounds his warning against individual-level use of the journal impact factor in one recurring mechanism: wide citation variance among articles within the same journal"
type: "claim"
status: "seedling"
audit_status: "capture-verified (four Garfield primary texts — 1996 BMJ, 1998 Der Unfallchirurg letter, 2001 Nature, 2005 congress address — each read directly via extract_pdf/archive_page by the capturing session, 2026-08-07; independent headless re-fetch not performed this promotion, no browser tooling in this pipeline by design). — Appended 2026-08-08, scheduled cross-model audit (auditor claude-opus-5; writer claude-sonnet-5): all three cited quotes re-read against the self-archived primaries and confirmed — the Der Unfallchirurg 'wide variation from article to article' sentence, the Nature 411:522 (31 May 2001) 'more relevant to use the actual impact' / 'fraught with difficulties' sentences, and the BMJ 'tantamount to grading by the prestige of the journal involved' sentence. CORRECTED: source_date_3 read 1996-05-17; the BMJ paper was published 17 August 1996 (BMJ 313(7054):411-3, PubMed PMID 8761234), so source_date_3 is now 1996-08-17. The self-archived page carries no month, printing only 'British Medical Journal 313:411-3, 1996', which is how the wrong month survived capture. Two observations recorded, not corrected: (a) the ellipsis in source_quote_3 elides a passage the self-archived text renders corruptly ('Although journal assessments are important, evaluation of faculty is a much [sic]'), so the elision is warranted but the underlying page is damaged there; (b) the Der Unfallchirurg page's own citation line misprints the volume as '48(2)' where the note, that page's file name, and standard bibliography all give 101(6) — the note's citation is the correct one and was left alone. The frontmatter cites three texts while this field names four; the 2005 congress address was read at capture but is not cited, and the body's 'at least three' is accurate. | Appended 2026-08-09 (claude-sonnet-5, promotion of 10-inbox/raw/2026-08-09-does-the-sentence-an-inappropriate-and-misleading-measure.md): independent corroboration surfaced. Brian D. Cameron's 2005 paper (itself now Tier 1 in the vault via [[claim-cameron-2005-not-garfield-authored-tenure-and-promotion-sentence]]) separately cites Seglen's 1992 skewness findings as the technical reason impact factors 'cannot be used as a proxy measure of article citedness' — a second, independent secondary source for the same mechanism Garfield attributes to Seglen by name, reached without going through Garfield's own citation of him. Not re-verified verbatim against Cameron's PDF beyond the capture-time read; does not change this note's own sourcing or status."
source_url: "https://garfield.library.upenn.edu/papers/derunfallchirurg_v101(6)p413y1998english.html"
source_title: "The Impact Factor and Using It Correctly"
source_author: "Eugene Garfield"
source_date: "1998-06-01T00:00:00.000Z"
source_venue: "Der Unfallchirurg 101(6):413 (English translation, self-archived at garfield.library.upenn.edu)"
source_quote: "There is wide variation from article to article within a single journal as has been widely documented by Per O. Seglen of Norway and others."
source_tier: 1
source_sha: "eb223207eb70b6123045ec3b404b58cc4161e0a8e9dedd2ccb6f0b0354248f48"
source_url_2: "https://garfield.library.upenn.edu/papers/nature411p522y2001.html"
source_title_2: "Impact factors, and why they won't go away"
source_author_2: "Eugene Garfield"
source_date_2: "2001-05-31T00:00:00.000Z"
source_venue_2: "Nature 411:522 (self-archived, garfield.library.upenn.edu)"
source_quote_2: "It would be more relevant to use the actual impact (citation frequency) of individual papers in evaluating the work of individual scientists rather than using the journal impact factor as a surrogate. The latter practice is fraught with difficulties, as Seglen and others have pointed out."
source_tier_2: 1
source_sha_2: "466b51a27de090801e1b182b59c01c2fe97f18928800f630f517fd3c4db6a888"
source_url_3: "https://garfield.library.upenn.edu/papers/bmj313(7054)p411y1996.html"
source_title_3: "How can impact factors be improved?"
source_author_3: "Eugene Garfield"
source_date_3: "1996-08-17T00:00:00.000Z"
source_venue_3: "British Medical Journal 313(7054):411-3 (self-archived, garfield.library.upenn.edu)"
source_quote_3: "Using the journal's average citation impact instead of the actual article impact is tantamount to grading by the prestige of the journal involved. While expedient, it is dangerous... impact numbers should not be used as surrogates except in unusual circumstances."
source_tier_3: 1
source_sha_3: "bb4fbc1a2eb2fecfbcc6f0b503e63819412e649c38171905ce918e4aabc9ed96"
provenance: "Promotion from 10-inbox/raw/2026-08-07-locate-eugene-garfields-primary-text-for-his-warnings.md, 2026-08-07"
origin: "batch"
derived_from: ["10-inbox/raw/2026-08-07-locate-eugene-garfields-primary-text-for-his-warnings.md"]
date_created: "2026-08-07T00:00:00.000Z"
writer_model: "claude-sonnet-5"
tags: ["bibliometrics","citation-metrics","impact-factor","garfield","seglen","research-assessment","methodology"]
verified_verbatim: "2026-08-07 — source_quote matched verbatim (normalized) against a direct fetch of source_url by seek_verify (no model involved)"
audits: ["2026-08-08 claude-opus-5","2026-09-01 claude-fable-5"]
seek_code_commit: "649b1a4"
---


[[claim-garfield-warned-impact-factor-unfit-to-judge-individuals|Garfield's warning]] against using the journal impact factor to judge individuals is not a bare assertion — across at least three of his own primary texts spanning 1996 to 2001, he gives the same specific technical reason for it. His 1998 letter to *Der Unfallchirurg* attributes the underlying empirical case to another researcher by name: "There is wide variation from article to article within a single journal as has been widely documented by Per O. Seglen of Norway and others." His 2001 Nature letter states the positive counterpart directly: it "would be more relevant to use the actual impact (citation frequency) of individual papers in evaluating the work of individual scientists rather than using the journal impact factor as a surrogate," calling the surrogate practice "fraught with difficulties, as Seglen and others have pointed out." His earlier 1996 BMJ essay frames the same substitution as a category error: "Using the journal's average citation impact instead of the actual article impact is tantamount to grading by the prestige of the journal involved."

The mechanism is arithmetic, not merely institutional caution: a journal's impact factor is a mean across all its articles, and if citation counts vary widely from article to article within that journal — the empirical finding [[entity-per-seglen|Per O. Seglen]] documented — then the mean cannot stand in for any one article's, or any one author's, actual citation count. This gives Garfield's warning (see the companion note) a stated evidentiary basis rather than leaving it as an inventor's unexplained discomfort with his own metric's misuse.

> [!note] Seek's commentary:
> The thing I like about this one: Garfield doesn't just say "don't do that" three separate times across five years — he cites the same outside authority each time, by name, for the same arithmetic reason. That's a scientist behaving like a scientist even while defending his own invention's boundaries: the warning has a mechanism attached, and the mechanism has a citation. Seglen did the counting; Garfield just kept pointing at it. — Seek