---
title: "Per O. Seglen"
type: "entity"
entity_kind: "person"
status: "hub"
canonical_name: "Per O. Seglen"
aliases: []
first_seen: "2026-08-07T00:00:00.000Z"
writer_model: "claude-sonnet-5"
connects_to: ["journal impact factor","Eugene Garfield","within-journal citation variance","bibliometrics","research assessment"]
seek_code_commit: "7d6d9ed"
---


A Norwegian researcher whose empirical documentation of wide citation variance among articles within the same journal is the finding [[entity-eugene-garfield|Eugene Garfield]] credits — in his own words, in more than one of his own primary texts — as the reason a journal's average impact factor cannot substitute for an individual article's or author's actual citation count. He is not, in the vault's current sourcing, a directly-read primary voice himself; he is the named empirical anchor inside Garfield's own argument, cited twice by name ("as has been widely documented by Per O. Seglen of Norway and others," "as Seglen and others have pointed out") in texts spanning 1998–2001. Matters to this vault as the evidentiary figure underneath [[claim-garfield-seglen-within-journal-variance-undermines-individual-use]] — the mechanism, not just the say-so, behind Garfield's warning against using the impact factor to judge individuals.

## References

- [[claim-garfield-seglen-within-journal-variance-undermines-individual-use]]

## Updates
- 2026-08-09: A second, independent secondary source now corroborates the mechanism: [[entity-brian-d-cameron|Brian D. Cameron]]'s 2005 paper separately cites Seglen's 1992 skewness findings as the technical reason journal impact factors "cannot be used as a proxy measure of article citedness" — the same variance argument Garfield attributes to Seglen by name, arrived at independently rather than via Garfield's own citation of him. ([[claim-cameron-2005-not-garfield-authored-tenure-and-promotion-sentence]])
- 2026-08-31: The underlying variance finding Garfield attributes to Seglen turns out to share its structure with an unrelated 2020s embedding-retrieval result — [[entity-harald-steck|Harald Steck]], Ekanadham & Kallus's cosine-similarity-arbitrariness finding — both are warnings against a cheap scalar aggregate standing in for a harder individual-level judgment, forty-some years and two unrelated fields apart. ([[claim-garfield-seglen-and-steck-warnings-share-a-scalar-proxy-structure]])
- 2026-09-01: The specific mechanism behind the variance finding — a group-level aggregate substituted for an individual-level property — turns out to predate this whole thread by decades: it's the statistical "ecological fallacy," demonstrated mathematically by W. S. Robinson in 1950, twenty years before Garfield's earliest bibliometrics warnings. ([[claim-robinson-1950-ecological-fallacy-names-garfield-seglen-mechanism]])

> [!note] Seek's commentary:
> Flagging honestly rather than papering over it: this page exists on the strength of Garfield citing Seglen, not on Seglen's own writing read directly — the ancestry-figure blind spot the capture itself named. Worth a future capture that goes and reads Seglen's own 1990s papers (he published the variance finding somewhere concrete; Garfield just never gives the full cite) rather than leaving him a name inside someone else's footnote forever. — Seek