<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Achilleas Samaras on Medium]]></title>
        <description><![CDATA[Stories by Achilleas Samaras on Medium]]></description>
        <link>https://medium.com/@achilleas.samaras?source=rss-23c4e783941c------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*8MR6un3yNcb9H3Oq0T29bA.jpeg</url>
            <title>Stories by Achilleas Samaras on Medium</title>
            <link>https://medium.com/@achilleas.samaras?source=rss-23c4e783941c------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 14:38:00 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@achilleas.samaras/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[A simple step to improve modern academia, revisited]]></title>
            <link>https://medium.com/@achilleas.samaras/a-simple-step-to-improve-modern-academia-revisited-08187cccbe06?source=rss-23c4e783941c------2</link>
            <guid isPermaLink="false">https://medium.com/p/08187cccbe06</guid>
            <category><![CDATA[research]]></category>
            <category><![CDATA[bibliometrics]]></category>
            <category><![CDATA[research-integrity]]></category>
            <dc:creator><![CDATA[Achilleas Samaras]]></dc:creator>
            <pubDate>Thu, 28 May 2026 23:48:16 GMT</pubDate>
            <atom:updated>2026-05-28T23:48:16.679Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Four years, one conference, one preprint, an empirical cohort, and an open-source tool later — the case for declared percentage contributions on every published paper.</em></p><p><strong>IN BRIEF</strong></p><p>Author-level research metrics — citation counts, <em>h</em>-indices, journal-impact aggregates — currently inherit paper-level numbers indiscriminately. A 600-citation paper with ten authors counts as 600 citations on each of their records, irrespective of who did the work. The proposal updated here closes a structural gap in scholarly metadata: at acceptance, <strong>co-authors declare a list of percentage contributions summing to 100%</strong>, <strong>and each declared share multiplies any paper-level metric to yield the author-level analogue. </strong>Paper-level numbers are preserved; an author-level set sits alongside them in a two-tier display. The full preprint, an open-source tool that exercises the metric arithmetic against real publication records, and a first empirical pass against a 100-researcher ocean-engineering cohort are linked at the end of this piece.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NncS0lUWfdYP2SkgOW1row.png" /></figure><p><strong>SIX AUTHORS IS THE NEW NORMAL</strong></p><p>In a random sample of 101,580 biomedical research papers uploaded to PubMed Central between 2016 and 2021, the median paper had six authors, and 90% of papers had between one and fifteen (Choueiry, 2022). Twenty years earlier, in 2002, the median was three. Single-authored papers, which made up about a third of the biomedical literature in 2002, were down to roughly 2% by 2021.</p><p><strong>The pattern is not confined to biomedicine.</strong> Thelwall and Maflahi (2022) document, across the entire Scopus-indexed literature, a steady rise in the geometric-mean number of authors per article: ~1.3 in the 1980s, ~1.5 in the 2000s, above 2 by the 2020s, with field-specific arithmetic means substantially higher across the life, health, and physical sciences. Immunology and microbiology — the most collaborative broad field they examine — now sits near six authors per paper on a geometric-mean basis. Larivière et al. (2015) extend the historical view back to 1900 and confirm the same trend; Abramo and D’Angelo (2015), looking specifically at Italian science, report comparable evidence. Mean citation impact has risen in step with team size across most disciplines for over a century, so the most cited work is increasingly the most multi-authored work.</p><p>Concurrently, the National Science Board’s Science and Engineering Indicators reports (NSB/NSF, 2024, 2025) put the annual count of peer-reviewed articles roughly tripling between 2003 and 2023, from ~1.22 million to ~3.28 million. <strong>More papers, more authors per paper.</strong></p><blockquote><strong>The bibliometric machinery built to assess this output was constructed for an earlier scale.</strong></blockquote><p><strong>WHAT EVALUATION ACTUALLY MEASURES</strong></p><p>Three operational metrics dominate academic evaluation: the number of papers a researcher has published, the impact factors of the journals they have published in, and the citations their papers have received. <strong>Each was originally defined as a paper-level number — a property of the paper, not of any one author.</strong></p><p>The trouble is that these paper-level numbers are then re-used, unchanged, as author-level signals.</p><blockquote><strong>A paper with one author and 600 citations contributes 600 citations to that author’s record. A paper with ten authors and 600 citations contributes 600 citations to each of the ten authors’ records.</strong></blockquote><p>Whatever the actual division of work, every co-author inherits the same number.</p><p>That cannot be the right rule when the median paper now has six authors. It also cannot be quietly corrected by the system as it stands: there is no piece of bibliographic metadata that records who did how much. The structural information needed to do better simply is not in the published record.</p><p>The major reform initiatives of the past decade — the San Francisco Declaration on Research Assessment (DORA, 2013), the Leiden Manifesto (Hicks et al., 2015), the Metric Tide review (Wilsdon et al., 2015), the Hong Kong Principles (Moher et al., 2020), and the Agreement on Reforming Research Assessment that underpins the global Coalition for Advancing Research Assessment (CoARA; Arentoft et al., 2022) — converge on a similar diagnosis: paper-level proxies, applied indiscriminately to authors, are an inadequate basis for evaluating researchers. None of these initiatives prescribes a specific numerical fix; all argue, at the level of principle, that one is needed.</p><p><strong>TWO GENUINE ATTEMPTS</strong></p><p>Two distinct strands of work do attempt something concrete.</p><p>The first is <strong>qualitative: the Contributor Roles Taxonomy (CRediT)</strong>, formalised as an ANSI/NISO standard in 2022 (NISO, 2022). CRediT defines fourteen contributor roles spanning conceptualisation, methodology, formal analysis, supervision, writing, and others. Most major publishers now require a CRediT statement at submission, and the statement appears alongside the published article. CRediT can record, in machine-readable form, who contributed to which aspect of the work. It also defines an optional “degree of contribution” attribute — <em>lead</em>, <em>equal</em>, or <em>supporting</em> — that may be attached to each role. The optional facet is rarely activated; the Editorial Manager platform, used by thousands of journals, has integrated it as a specific configuration that most journals have not enabled (Holcombe et al., 2020). The output is a list of roles, not a list of shares.</p><p>The second is <strong>quantitative: the counting-methods literature</strong>. Gauffriau (2021) catalogues twenty-nine distinct counting methods introduced between 1970 and 2018. Full counting, in which each author of an <em>n</em>-author paper receives one credit. Fractional counting (Van Hooydonk, 1997), in which each receives 1/<em>n</em>. Harmonic counting (Hodge and Greenberg, 1981; Hagen, 2008, 2010), in which the <em>i</em>-th author of <em>n</em> receives a credit proportional to 1/<em>i</em> normalised across the list. Geometric, arithmetic, first-author-only, last-author-only, and field-modified variants. Hagen (2010) reports that harmonic counting fits self-reported contribution shares from psychology, medicine, and chemistry more closely than the alternatives.</p><p>Every method in this family <strong>imputes</strong> the share algorithmically, from author position or author count or both. <strong>None lets the co-authors themselves declare what they actually did.</strong></p><p><strong>THE PROPOSAL IN ONE PARAGRAPH</strong></p><p>The proposal is that at final acceptance — alongside the CRediT statement, after peer review — <strong>the co-authors of a published paper declare a list of percentage contribution shares summing to 100%</strong>. Each declared share then multiplies any paper-level metric — citation count, journal impact factor, altmetric score, or any other — to yield the corresponding author-level analogue for that contributor. Paper-level metrics are preserved unchanged. Each researcher’s bibliometric profile is then expressed as <strong>two complementary metric sets</strong>: a paper-level set, identical to current practice and representing each work’s contribution to its field; and an author-level set, novel to the proposal, representing the individual’s declared contribution. <strong>Implementation requires no new infrastructure</strong>: one validated numeric field per author, added to the existing CRediT submission workflow, stored in the article’s JATS XML metadata, and propagated to bibliometric databases through the same channels currently used for the qualitative CRediT statement.</p><p>The first version of this idea was sketched in a short <a href="https://medium.com/@achilleas.samaras/a-simple-step-to-improve-modern-academia-33c56288158f">Medium piece in January 2022 (Samaras, 2022)</a> and presented orally at the 8th World Conference on Research Integrity in Athens (Samaras, 2024). The full formal treatment — the case studies, the implementation specification, the state-of-the-art discussion, and the empirical exercise that the next two sections summarise — is in a 2026 MetaArXiv preprint (Samaras, 2026; <a href="https://doi.org/10.31222/osf.io/3mkz2_v1">DOI</a>).</p><p><strong>WHAT THE SYSTEM DOES, BY ARCHETYPE</strong></p><p>To see how the proposal would re-rank researchers, the preprint constructs <strong>six archetypal profiles</strong>, organised by contribution style and intentionally decoupled from career stage. Each is summarised by a constructed publication list with plausible authorship positions, co-author counts, and citation counts, from which the paper-level and author-level totals are computed under the proposed system. The numbers below are order-of-magnitude estimates from those constructed lists, intended to expose the proposal’s <em>behaviour</em> across contribution styles, not to claim anything about how common each style is in any field.</p><p><strong>- Profile A — Solo / lead-author specialist.</strong> Small groups (1–3 co-authors), consistently lead or sole position. Declared shares 60–100%. The two metric sets stay close throughout the record. Ranking improves relative to peers whose raw counts are inflated by peripheral co-authorships.</p><p><strong>- Profile B — Small-group equal collaborator.</strong> Two- or three-person partnerships in which the work is performed and declared as equal. Shares 30–50%. Author-level total ~38% of paper-level. Neither set alone tells the full story; the two-tier display is particularly informative.</p><p><strong>- Profile C — Balanced mid-size collaborator. </strong>Groups of 4–7, varied positions, shares 20–40%. Author-level total ~30% of paper-level. The honest broad collaborator is not penalised — the contribution is simply re-attributed proportionally, and the paper-level set continues to reflect the full collaborative footprint alongside.</p><p><strong>- Profile D — Supervisory contributor. </strong>Predominantly last-author, larger groups, sustained output over a long active period. Shares 15–25%. Author-level total ~21% of paper-level. Ranking remains substantial; sustained mentorship is preserved as a recognised contribution.</p><p><strong>- Profile E — Consortium contributor. </strong>Frequent co-authorship on large-collaboration papers (10–100+ authors). Shares ≤2% on most consortium papers, often combined with a smaller number of papers in smaller groups. The author-level total collapses on the consortium papers but accumulates normally on the small-group work, which becomes proportionally visible.</p><p><strong>- Profile F — Honorary / gift author. </strong>Persistent low shares (5–10%) across small- and medium-sized groups where a higher contribution would be expected. Author-level total ~8% of paper-level. Profile F is structurally distinct from Profile E: the consortium contributor’s small shares are legitimate small contributions to genuinely large collaborations, while the honorary author’s small shares occur where they are not. The two-tier display makes that distinction visible as a <em>pattern</em> across the record — a large paper-level set paired with a persistently small author-level set, in the absence of consortium-paper structure, is the bibliometric signature of gift authorship.</p><p>The summary comparison table sits inside the preprint. The proposal’s effect, viewed across the archetypes, is not a uniform discount or reward — it is a redistribution of attribution along the actual contribution structure of the work.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*j-QcYj6eSmFyqs9p1u6M7Q.png" /></figure><p><strong>WHAT IT LOOKS LIKE ON A REAL FIELD</strong></p><blockquote><strong>The proposal calls for declared shares. Declared shares do not yet exist anywhere in the published literature, because no journal requires them.</strong></blockquote><p>To test the proposal’s behaviour against a real publication record, the work above is accompanied by an <strong>open-source tool — “dpac”</strong>, <em>declared percentage-based author-contribution</em> (Samaras, 2026) — that computes the two-tier metrics for any researcher’s OpenAlex record. In place of the declared shares the proposal calls for, the tool uses two algorithmic stand-ins: equal-share (1/<em>n</em>) and harmonic counting. These are well-understood, scheme-agnostic substitutes; they are not a replacement for declared shares, only a way to exercise the metric arithmetic on real data.</p><p>A first run took a stratified random sample of 100 active ocean-engineering researchers, journal articles only. Under harmonic counting, the median ratio of author-level to paper-level citations was <strong>0.30</strong>, with an interquartile range of [0.20, 0.37] and a full range of 0.10–0.66. Under equal-share, the median was 0.27 (IQR [0.21, 0.33], range 0.10–0.49) — tighter, because equal-share does not reward lead-author positions and so does not stretch the upper tail.</p><p>Of the 100 researchers, <strong>76 fell within the central band between Profile B and the F/D pair</strong> under harmonic counting (87 under equal-share). Eighteen sat above this band toward Profile A; six sat below toward Profile E. Profiles F and D themselves form the band’s lower edge: they overlap visually under harmonic counting and sit adjacent under equal-share, so the F/D distinction cannot be made on ratio alone in either case — separating the two requires looking at the pattern across a record, not the single number. Profile E — the consortium contributor — sat well outside the populated cloud, as expected: ocean engineering does not produce consortium-scale author lists, and Profile E is the field-specific reference marker that cross-field comparisons with clinical medicine, mathematics, or high-energy physics would be expected to populate.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0CiBC17_U4X8LGpuyGmx3A.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iljSXWJwCXK3Jq8J_ft0TA.png" /></figure><p>That is, in a single field, a single empirical pass: most active researchers’ author-level citation totals come out at roughly a quarter to a third of their paper-level totals once the contribution share is taken into account; the archetypes function as reference markers on a continuum rather than as discrete bins; and the proposal differentiates contribution styles in ways that map onto recognisable researcher profiles. The full plots, code, and demonstration notebook are in the <strong>“dpac” repository</strong> (<a href="http://github.com/achilleassamaras/dpac">github.com/achilleassamaras/dpac</a>; Zenodo concept <a href="https://doi.org/10.5281/zenodo.20359542">DOI</a>).</p><p><strong>WHAT THIS DOES NOT SOLVE</strong></p><p><strong>The proposal is deliberately scoped to address the attribution problem — the indiscriminate transfer of paper-level numbers to every co-author — and to do so with a single, well-defined intervention.</strong> It does not address misuse of the journal impact factor in researcher evaluation; that is DORA’s territory. It does not address citation cartels, salami slicing, or other gaming behaviours; those are integrity-and-incentives questions of their own. It does not propose a new master metric to replace the <em>h</em>-index; for rank-based indices, the obvious extension is the fractional <em>h</em>-index of Egghe (2008), and the preprint recommends it, but the design remains agnostic about which paper-level metric the percentage shares are applied to.</p><p>What the proposal does is supply the <strong>concrete computational layer that the major reform initiatives — DORA, the Leiden Manifesto, the Hong Kong Principles, CoARA — have called for at the level of principle but have not specified</strong>. It is downstream-compatible with all of them. It takes the existing CRediT submission workflow and extends it from an optional, qualitative, three-level ordinal to a mandatory quantitative percentage suitable for downstream metric computation. The structural information needed is generated at submission, declared by the people best positioned to declare it, and propagated through the same publishing pipeline that already carries everything else.</p><p><strong>WHERE THIS STANDS</strong></p><blockquote><strong>A perfect system might never exist; the current one can and should be improved.</strong></blockquote><p><strong>The case for that improvement is, four years on from the initial sketch, stronger than ever. </strong>The median paper is bigger than ever. The bibliometric apparatus underneath is the same one that has been used since the 1970s. The closest existing standard — CRediT — is already in place at most major publishers. <strong>The change is small to implement and large in what it would surface:</strong> a CRediT field extended from optional ordinal to mandatory percentage, propagated through the publishing pipeline that already carries everything else, and the attribution structure of a century’s worth of multi-authored work made visible for the first time. The empirical work to demonstrate the proposal’s behaviour against a real field is now done.</p><p>The full preprint, with the formal definition, the state-of-the-art discussion, the case studies, and the empirical exercise summarised above, is on MetaArXiv (<a href="https://doi.org/10.31222/osf.io/3mkz2_v1">DOI</a>). The open-source tool, with the demonstration notebook that reproduces the cohort plot, is on GitHub at <a href="http://github.com/achilleassamaras/dpac">github.com/achilleassamaras/dpac</a> and archived on Zenodo at <a href="https://doi.org/10.5281/zenodo.20359542">DOI</a>. The original 2022 sketch and the 2024 WCRI presentation are archived on Zenodo at <a href="https://doi.org/10.5281/zenodo.20355359">DOI</a> and <a href="https://doi.org/10.5281/zenodo.20355558">DOI</a>, respectively.</p><p><strong><em>FURTHER READING AND SOURCES</em></strong></p><p><em>Preprint and companion tool</em></p><p>- Samaras, A. G. (2026). A declared percentage-based author-contribution system for weighted research metrics: a proposal. MetaArXiv. DOI: <a href="https://doi.org/10.31222/osf.io/3mkz2_v1">https://doi.org/10.31222/osf.io/3mkz2_v1</a><br>- Samaras, A. G. (2026). “dpac”: Declared percentage-based author-contribution metrics from OpenAlex publication records (v1.0.0). Open-source software (Python). <a href="https://github.com/achilleassamaras/dpac">https://github.com/achilleassamaras/dpac</a> · Zenodo concept DOI: <a href="https://doi.org/10.5281/zenodo.20359542">https://doi.org/10.5281/zenodo.20359542</a></p><p><em>Prior versions of this proposal</em></p><p>- <a href="https://medium.com/@achilleas.samaras/a-simple-step-to-improve-modern-academia-33c56288158f">Samaras, A. G. (2022). A simple step to improve modern academia. Medium, 15 January</a>. Self-archived on Zenodo: <a href="https://doi.org/10.5281/zenodo.20355359">https://doi.org/10.5281/zenodo.20355359</a><br>- Samaras, A. G. (2024). The case for a percentage-based author-contribution system in scientific publishing. 8th World Conference on Research Integrity, Athens, Greece, 2–5 June 2024 (OP33.1). Archived on Zenodo: <a href="https://doi.org/10.5281/zenodo.20355558">https://doi.org/10.5281/zenodo.20355558</a></p><p><em>Authorship trends and bibliometric scale</em></p><p>- Choueiry, G. (2022). Does the Number of Authors Matter? Data from 101,580 Research Papers. Quantifying Health (blog). <a href="https://quantifyinghealth.com/number-of-authors-of-research-papers/">https://quantifyinghealth.com/number-of-authors-of-research-papers/</a><br>- Thelwall, M., and Maflahi, N. (2022). Research coauthorship 1900–2020: continuous, universal, and ongoing expansion. Quantitative Science Studies, 3(2), 331–344. DOI: 10.1162/qss_a_00188<br>- Larivière, V., Gingras, Y., Sugimoto, C. R., and Tsou, A. (2015). Team size matters: collaboration and scientific impact since 1900. Journal of the Association for Information Science and Technology, 66(7), 1323–1332. DOI: 10.1002/asi.23266<br>- Abramo, G., and D’Angelo, C. A. (2015). The relationship between the number of authors of a publication, its citations and the impact factor of the publishing journal: evidence from Italy. Journal of Informetrics, 9(4), 746–761. DOI: 10.1016/j.joi.2015.07.003<br>- NSB/NSF (2024). Science and Engineering Indicators 2024: The State of U.S. Science and Engineering (NSB-2024–3). National Science Board / National Science Foundation, Alexandria, VA.<br>- NSB/NSF (2025). Discovery: R&amp;D Activity and Research Publications, Science and Engineering Indicators 2026 (NSB-2025–7). National Science Board / National Science Foundation, Alexandria, VA.</p><p><em>Reform initiatives</em></p><p>- DORA (2013). San Francisco Declaration on Research Assessment.<br>- Hicks, D., Wouters, P., Waltman, L., de Rijcke, S., and Rafols, I. (2015). The Leiden Manifesto for research metrics. Nature, 520, 429–431.<br>- Wilsdon, J. et al. (2015). The Metric Tide: Report of the Independent Review of the Role of Metrics in Research Assessment and Management.<br>- Moher, D., Bouter, L., Kleinert, S., Glasziou, P., et al. (2020). The Hong Kong Principles for assessing researchers: fostering research integrity. PLOS Biology, 18(7), e3000737.<br>- CoARA / Arentoft, M., et al. (2022). Agreement on Reforming Research Assessment. Coalition for Advancing Research Assessment.</p><p><em>Contributorship and counting method</em>s</p><p>- NISO (2022). Contributor Roles Taxonomy (CRediT) Standard. ANSI/NISO Z39.104–2022.<br>- Holcombe, A. O., Kovacs, M., Aust, F., and Aczel, B. (2020). Documenting contributions to scholarly articles using CRediT and tenzing. PLOS ONE, 15(12), e0244611.<br>- Gauffriau, M. (2021). Counting methods introduced into the bibliometric research literature 1970–2018: a review. Quantitative Science Studies, 2(3), 932–975.<br>- Hodge, S. E., and Greenberg, D. A. (1981). Publication credit. Science, 213(4511), 950.<br>- Hagen, N. T. (2008). Harmonic allocation of authorship credit: source-level correction of bibliometric bias assures accurate publication and citation analysis. PLOS ONE, 3(12), e4021.<br>- Hagen, N. T. (2010). Harmonic publication and citation counting: sharing authorship credit equitably — not equally, geometrically, or arithmetically. Scientometrics, 84(3), 785–793.<br>- Karthik, V., Anand, I. S., Mahanta, U., and Sharma, G. (2025). Authorship-contribution normalized Sh-index and citations are better research output indicators. arXiv preprint 2509.04124. DOI: 10.48550/arXiv.2509.04124<br>- Van Hooydonk, G. (1997). Fractional counting of multiauthored publications: consequences for the impact of authors. JASIS, 48(10), 944–945.<br>- Egghe, L. (2008). Mathematical theory of the h- and g-index in case of fractional counting of authorship. JASIST, 59(10), 1608–1616.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=08187cccbe06" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A simple step to improve modern academia]]></title>
            <link>https://medium.com/@achilleas.samaras/a-simple-step-to-improve-modern-academia-33c56288158f?source=rss-23c4e783941c------2</link>
            <guid isPermaLink="false">https://medium.com/p/33c56288158f</guid>
            <category><![CDATA[integrity]]></category>
            <category><![CDATA[scientific-publications]]></category>
            <category><![CDATA[science]]></category>
            <category><![CDATA[academia]]></category>
            <category><![CDATA[research]]></category>
            <dc:creator><![CDATA[Achilleas Samaras]]></dc:creator>
            <pubDate>Sat, 15 Jan 2022 12:59:29 GMT</pubDate>
            <atom:updated>2022-01-15T12:59:29.998Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*91rDrdACaL6Ljzwl_yuf_A.jpeg" /><figcaption>Photo by <a href="https://www.pexels.com/@thepoorphotographer?utm_content=attributionCopyText&amp;utm_medium=referral&amp;utm_source=pexels">Janko Ferlic</a> from <a href="https://www.pexels.com/photo/light-inside-library-590493/?utm_content=attributionCopyText&amp;utm_medium=referral&amp;utm_source=pexels">Pexels</a></figcaption></figure><p>People involved in modern academia are well aware of its problems. If you are new to academia or unaware of these problems, there are places to start before reading this article (<a href="https://www.liebertpub.com/doi/10.1089/ees.2016.0223">Link</a>).</p><p>Regardless of awareness, though, there is little to no action taken. The reasons why are well known — even if rarely outspoken — and listing them exceeds the scope of this article.</p><p>In the paragraphs that follow I will focus on scientific publications (i.e. peer-reviewed journal publications), which play a key-role in most stages of career development in modern academia. This role can be summed-up to three “axioms”:</p><ol><li>The more in number, the better.</li><li>The higher the journal impact, the better.</li><li>The higher the citation count, the better.</li></ol><p><strong>In theory this should work. Right?</strong></p><p><strong>Yet, it doesn’t. </strong>At least not today, in the overpopulated, highly-competitive, “perform-according-to-some-standard”-or-perish environment of academia.</p><p>Again, I will spare you the details about the reasons why and the ways how many people “hack the system” and exploit it for personal gain. Instead, I will go straight to the simple step I promised in my title.</p><blockquote><strong><em>A percentage-based author-contribution system in scientific publishing.</em></strong></blockquote><p>In an attempt to avoid overpopulated authors’ lists and to promote integrity, many journals nowadays require that the authors add to the end of their manuscript a declaration of authors’ contributions.</p><p>This is a good idea but doesn’t solve the problem, as: (a) only a few people actually read and/or are interested in this declaration; and (b) this declaration does not actually affect the added value each author gets from the published article.</p><p>The new system proposed here is based on the same rationale, but with one very important modification: <strong>co-authors will have to declare (and thus agree to) a percentage-based author-contribution list, </strong>totaling to 100% (since the article represents a unit itself).</p><p>Following this, all research output by an author and future evaluations should be judged accordingly.</p><p>Need an <strong>example</strong>?</p><p>Let us consider two research groups: Group A, consisting of 2 researchers and Group B, consisting of 5 researchers. In the hypothetical scenario where Groups A and B publish concurrently (in parallel universes probably) the exact same paper in the same journal, and they do so having all of their members contribute equally to the scientific output, Group A researchers will be each attributed with 50% of their <strong>contribution to the field</strong>, while Group B researchers will be each attributed with 20% of their <strong>contribution to the field</strong>.</p><p>The term in bold is a key-aspect in this concept, as <strong>this contribution can be associated with whichever metric</strong> (journal impact, citations, etc) and thereafter used for comparative evaluations.</p><p><strong>Taking our example one step further</strong>, we now have a scenario in which a committee needs to blindly evaluate for a position two candidates based on their contribution to their field: one researcher from Group A and one from Group B. If their previously defined “identical” article is their sole research output, isn’t it clear who should get the position? And this, regardless of the metric attached to their contribution. Should this be journal impact, number of citations, combination of the two, or any other devised criterion, 50% will always beat 20%. Does the approach sound unfair to you? I hope not.</p><p>Based on the above, it is easily deduced how this approach could be <strong>expanded to all other aspects of author-related metrics</strong> and how seamlessly this could be done. Its implications for comparative evaluations and the bad practices it will harm are easily deduced as well.</p><p>When theorising about the implementation of this approach, I have carefully considered a handful of issues one might raise against it; and have come to address them all to a more than adequate extent, as well. I see no need to tire you with lengthy arguments at this stage (the numerical examples of Tables 1, 2 and 3 could provide some insights); I would urge you, though, to <strong>answer to three simple questions</strong>:</p><ol><li>Do you disagree with the informed use of quantitative metrics in research/academia?</li><li>Does the presented approach alter the value of each scientific publication as a single, unique contribution to its field?</li><li>Does this approach favour the people exploiting inadequacies of the present system?</li></ol><blockquote><em>If your answers to all three questions are </em><strong><em>“No”</em></strong><em>, then we’re on the same page.<br>It sounds only reasonable to push for this simple step.</em></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/388/1*RJaCGmSGOz7NC2qLIg25vA.png" /><figcaption>Table 1. How a percentage-based author-contribution system would address typical issues on the evaluation of author-related metrics (<em>contribution impact evaluated on the basis of citation count and H-index; grey=without the new approach, black=with the new approach</em>).</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/720/1*8g0oV6yasdRQrxLfn5Xgyw.png" /><figcaption>Table 2. How a percentage-based author-contribution system would address typical issues on the evaluation of author-related metrics (contribution impact evaluated on the basis of journal impact; grey=without the new approach, black=with the new approach).</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/720/1*7-Cguho0aZCAmCHRlLi9pg.png" /><figcaption>Table 3. How a percentage-based author-contribution system would address typical issues on the evaluation of author-related metrics (contribution impact evaluated on the basis of journal impact combined with citation count; grey=without the new approach, black=with the new approach).</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=33c56288158f" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>