Wikidata:Property proposal/Impact factor

impact factor

Originally proposed at Wikidata:Property proposal/Generic

   Not done

Description	scientometric index calculated by Clarivate that reflects the yearly average number of citations of articles published in the last two years in a given journal
Represents	impact factor (Q5330)
Data type	Number (not available yet)
Template parameter	impact (in Infobox journal on enwiki)
Domain	academic journal (Q737498)
Example 1	Reviews in Medical Virology (Q15766137) → 4.221 point in time (P585) 2019
Example 2	Nature (Q180445) → 42.778 point in time (P585) 2019
Example 3	MISSING

Motivation

I'm currently interested in listing papers about ivermectin given that the topic is controversial. While listing the papers I want to show the impact factors of the journals in which the papers are published, so that a reader can easily go to the papers that are published in the journals with the highest impact factors. Currently, we lack a property that captures the impact factor.

Given that it's a value that gets used in Wikipedia infoboxes it's also possible that they source it from Wikidata in the future. ChristianKl ❪✉❫ 11:29, 29 June 2021 (UTC)

Note: This would be a restoration of P1169 (P1169).--GZWDer (talk) 16:18, 29 June 2021 (UTC)

Discussion

ChristianKl ❪✉❫ 11:29, 29 June 2021 (UTC)

See:

• Wikidata:Property_proposal/Archive/19
• Wikidata:Requests for deletions/Archive/2014/Properties/1
• Wikidata:Property proposal/Archive/46

--GZWDer (talk) 12:37, 29 June 2021 (UTC)

• I don't see how the case is different from the usage in Wikipedia. As far as database rights goes, I have the impression that it's something by which EU databases are protected and the Thompson Media Group seems to be from the US. ChristianKl ❪✉❫ 15:28, 29 June 2021 (UTC)
•  Oppose The copyright issue seems serious, though the owner of IF has changed since the deletion request so perhaps the status can be reviewed, or they can be queried directly? There is also DORA - "General Recommendation: Do not use journal-based metrics, such as Journal Impact Factors, as a surrogate measure of the quality of individual research articles, to assess an individual scientist’s contributions, or in hiring, promotion, or funding decisions." ArthurPSmith (talk) 17:49, 29 June 2021 (UTC)
  • @ArthurPSmith: Wikipedia does list impact factors, so there are years without it being a problem. What law do you think protects this data? I would love to live in a world where people can be convinced based on the merits of arguments, but that's not our world. Sometimes it's required to argue that a study does have authority because it was published in a reputable journal. I would love for someone to do a page-rank style algorithm based on Wikidata's citation data but unfortunately that's not available currently. ChristianKl ❪✉❫ 23:34, 5 July 2021 (UTC)
• Why would you want to sort controversial articles based on IFs of the journal? These two things are not related. --Egon Willighagen (talk) 19:20, 29 June 2021 (UTC)
  • @Egon Willighagen: IF's are a way to identify no name journals that someone in India made where everyone can publish whatever they want and distinguish them for reputable journals. It's not a perfect way but it works. ChristianKl ❪✉❫ 23:34, 5 July 2021 (UTC)

    There are much easier ways to do that. The IF is not only not perfect, causes a lot of damage too. --Egon Willighagen (talk) 14:52, 6 July 2021 (UTC)

    See also Prestigious Science Journals Struggle to Reach Even Average Reliability (Q51430192) --Egon Willighagen (talk) 14:54, 6 July 2021 (UTC)

• While metrics of journal citations, quality or impact are only peripherally useful as metrics of the quality of a specific journal article, they do help to flag predatory publications and to red flag articles therein. When deciding whether a specific journal article should be treated as a [WP:RS] for an assertion contrary to “common knowledge” one looks for some objective indicator that the article has been through meaningful peer review by subject matter experts. Articles published in journals with few expert reader-correspondents might still be seen as credible if their authors are themselves highly regarded and can be shown to be so, but that is a level of examination rarely undertaken. Holding the line on pseudosciences being pushed into articles is a challenge. It would of course be preferable to have open, objective metrics of author reliability, but I don’t think we are there yet. In the mean time the I.F. and its kin still have some value. That said, the copyright question must be resolved. LeadSongDog (talk) 18:41, 8 July 2021 (UTC)
• late to the party here, but oppose. Impact factors are a proprietary measure developed by a single company (JCR, currently owned by Clarivate) and are accessible only by subscription. I do have access to JCR via my university library; I just reviewed the Clarivate terms of service which read: "Users may on an infrequent, irregular and ad hoc basis, distribute limited extracts of our data as incidental samples or for illustrative or demonstration purposes in reports or other documentation created in the ordinary course of their role. We determine a ‘limited extract’ as an amount of data that has no independent commercial value and could not be used as a substitute for any service (or a substantial part of it) provided by us, our affiliates or third party providers." So as before in the deletion discussion for this property - the company allows use on single webpages, but creating a database of impact factors, as we would with Wikidata, would violate the terms of use and individual license. While I would love to see impact factors opened up, they are also a highly imperfect measure, and aren't a good analogue for quality. Also, just in practical terms, they change often, so would require extremely frequent updating. Pinging @megs: too in case I'm misreading something in the Terms. -- Phoebe (talk) 20:13, 22 August 2021 (UTC)
  • I don't mean to reopen the vote, but @Phoebe: guess you're right about JCR IF. However, there are many other impact factors calculated by various bodies, and many are openly available. Eg there's h-index and i10-index on Google Scholar, and use that often for a quick check on some academic person --Vladimir Alexiev (talk) 08:07, 16 September 2022 (UTC)
•  Not done - no consensus to create this property --DannyS712 (talk) 01:09, 2 January 2022 (UTC)