Consider adding a gender aspect

[Daniel-Mietchen]

alternatively, we could could add more gender-based panels (like this one%20%3Fgender%20WHERE%20%7B%0A%20%20%20%20%3Fwork%20wdt%3AP1433%20wd%3AQ1981225%20.%0A%20%20%20%20%3Fwork%20wdt%3AP50%20%3Fauthor%20.%0A%20%20%20%20%3Fauthor%20wdt%3AP21%20%3Fgender%20.%0A%20%20%7D%20%0A%20%20GROUP%20BY%20%3Fgender%0A%7D%20AS%20%25result%0AWHERE%20%7B%0A%20%20INCLUDE%20%25result%0A%20%20SERVICE%20wikibase%3Alabel%20%7B%20bd%3AserviceParam%20wikibase%3Alanguage%20%22en%2Cda%2Cde%2Cep%2Cfr%2Cjp%2Cnl%2Cno%2Cru%2Csv%2Czh%22%20.%20%7D%20%0A%7D%20%0A%20ORDER%20BY%20DESC(%3Fcount)%0A%20) from the venue aspect) to the other aspects (e.g. to publisher or country)

[Daniel-Mietchen]

or we could go for a diversity aspect, which could have panels on different aspects of diversity and slice them in different ways (e.g. by institution/ topic/ journal/ country etc.).

[fnielsen]

There is one for the property: https://tools.wmflabs.org/scholia/P21

[prototyperspective]

I strongly oppose this as it amplifies data that and unnecessarily facilitates work and calls for unscientific and unethical, harmful, unfair, opinion-based interventions in the scientific process / mechanisms. Some relevant details are at #442.

Artificially trying to optimize for some opinion-based view of how "diverse" a "institution/ topic/ journal/ country" should be reduces efficiency, reduces effectiveness and increases unfairness.

If you're looking to improve things in that regard you could add information about projects to increase STE-related interests, or information which problems get how much science-funding or information about how much journals charge (article processing charges). Things like that would be useful and ethical.

[fnielsen]

For inspiration, there are some gender-related panel wrt. Danish films here: https://fnielsen.github.io/hack4dk2018/

[fnielsen]

For NeuroImage journal based on query by @Daniel-Mietchen and on https://fnielsen.github.io/hack4dk2018/

• Bar chart
• Line chart

[fnielsen]

For the NIPS/NeurIPS series:

• Line chart

[fnielsen]

Interesting pattern for the Nature journal, where 1974-1979 shows a bump and then a drop.

[fnielsen]

@prototyperspective I have a tendency to think that aggregated statistics are interesting. I do not think that such statistics lead journals to introduce gender-based quota - as an example.

Article processing charge statistics is indeed interesting. Unfortunately, that is not (yet) recorded in Wikidata, AFAIK. Or is it?

[prototyperspective]

They can be interesting but in practice having these statistics this visible is I think counterproductive and undue. I don't think it's easy to find substantial insights this way and it facilitates misinterpretations and so on. Instead of having this panel on all the pages by default, I'd support having a separate page about gender aspects where things like the listed examples are included and one can easily build or use such queries (e.g. the queries currently used on all the pages) and maybe some info in some documentation about this (in a FAQ-way) if there is some. (And I think your examples are interesting but there isn't any significant conclusion that can be drawn from them even if the data-quality and -quantity is sufficient.)

I don't think APCs is in Wikidata, that would be a separate issue and after posting I noted it down for revisiting this later and researching if it's possible, whether it would be useful and where/how to propose this. It wouldn't just be statistics but also a way to compare journals. I think these vary for each journal (each journal has many different APCs depending on various things, not just one) and over time. A lot of this data is public and I guess it would need to be proposed at Wikidata or at least the issue would involve first adding this data to the database.