mastodon 4.7.3Prof of Research Communications and co-lead of @COKIProject - http://orcid.org/0000-0002-0068
Fascinating:
Australian universities go up in QS rankings because of a change of methodology and this is a great endorsement of progress in the sector.
Australian universities go down in the THES rankings due to a change in methodology (and geopolitical change outside of our control) and it is a major disaster and concern.
These happen within three months of each other with virtually no reporting connecting the dots that this provides zero evidence of any actual change...
The COKI team are excited excited to be collaborating with @cwtsleiden to deliver the short course Scientometrics Using Open Data. 31 Oct - 3 Nov. Join us! Read all about it here: https://www.cwts.nl/education/cwts-course-program/scientometrics-using-open-data
#ResearchAssessment #OpenData #CoARA
@ludowaltman@akademienl.social @MsPhelps@akademienl.social @LucyMontgomery_@toot.community @neesjanvaneck@akademienl.social
It was really great to host @jameswilsdon@mstdn.social in Perth with COKI last week. In an effort to get down some general thoughts and concerns on Responsible Research Evaluation (and Australia's ranking obsessions) we wrote this for Campus Morning Mail:
https://campusmorningmail.com.au/news/rankings-tail-wags-the-strategy-dog/
#ResponsibleResearchAssessme #rra #era #OpenData #coara #rankings
Wondering what is going on post-ERA? Puzzled that combining WA unis could possibly increase rankings?
We're hosting world-leading expert @jameswilsdon@mstdn.social in Perth next week and you can hear him on the future of research evaluation on Wednesday 8th at 5pm
Really great to see the @libscie@akademienl.social update on #POSI This is a great example of using principles critically to think about how to move forward and critically assessing progress
@mpe@ravenation.club There's a poor taste joke in there somewhere about the wisdom of running Airflow on a dialysis machine, but I'll let it bubble away in the background...
@judell@social.coop To be fair the answer to that probably depends on which end of Market your head office is at...
If you prefer your #ResearchEvaluation #ResponsibleResearchAssessment rants in video form rather than written then this video interview with myself and @jameswilsdon@mstdn.social might be what you're looking for (link will autoplay):
https://www.futurehe.com.au/expert-opinion/?playlist=11c8b61&video=ab7d976
@hedgielib@glammr.us I bet it ends...
"..and the ethics approval was coming from /inside/ the research group..."
@hedgielib@glammr.us "We'll just get the data scientists to handle all our data management..." sounds like the start of one of those micro-horror stories...
@m@social.tthi.as You might think that with everyone concerned about Environment and Impact evaluation as a core piece of the future for Australian HE and the Accord process being concerned about wider impacts that APO would be a natural focus for attention and resources...
@jelena3121@mastodon.social Absolutely yes, I think the timeframe question is important. The BookSprint methodology makes a virtue of the constraint of time which creates some limitations. We've argued with the BS team that they should adapt it to allow for it being the start of a more open ended process rather than a bounded one with a fixed end point product.
Community/collaboration building vs output creation I guess
@MsPhelps@akademienl.social @brembs@mastodon.social
yes would be worth doing that time-domain analysis against zero embargo repository OA, at least where we've got data to allow it...fiddly tho, I need to template the query I guess...
@deradam@openbiblio.social Hmmm, its a bit difficult to tell from the screenshot. It would be best to go back to the data archives and check I think...
I've got reasonably used to tracking the gradual improvement in embargoed repository OA. This looks different to me but I don't think I can clearly articulate why, hence my question looking for orthogonal evidence.
My main counter is the UK - earlier agreements and a significant flat lining over the past several years. TAs more recent in most other places...
@neilernst@mastodon.acm.org Well Friday did improve with actual coffee so there is that...
My coffee tasted foul and weak.
Then I remembered I hadn't actually ground any coffee to put in the head. Must have reused yesterdays spent grounds.
How's your Friday going?
@williamgunn@mastodon.social In the UK, NL, Sweden, Germany and Australia there has been no increase in overall OA (unpaywall.is_oa) in the past 3 years (5 years for the UK) roughly coinciding with mass adoption of R&P in this countries
It's hard to disentangle embargoes entirely but this doesn't look like that
OA publishing goes up, but "green only" goes down so overall OA is roughly static
@williamgunn@mastodon.social Well we can debate the relative importance but see my first point.
It appears we're not making overall progress on access despite spending a lot of money. And that would mean no progress on STEM either
Observation from Data: I'm not seeing any overall increase in OA in those countries heavily invested in R&P agreements despite increases in OA publishing
Direct Consequence: that would mean that OA mediated exclusively through other platforms is decreasing
My Concern: attention (and money) focused on R&P deals is taking away from time and energy for tackling the remaining hard problems in access (of which there are several types, HSS is just one)
@jelena3121@mastodon.social Have looked at/been involved in book sprints (as specifically run by https://www.booksprints.net/)?
There's a bunch of parallels in the approach they bring to building a book by consensus in a short timeframe that mirrors some of what you're describing
And that approach is my general goto if I wanted to bring a group together to do (I thnk!) what you're describing as theorizing together (eg https://direct.mit.edu/books/oa-monograph/5131/Open-Knowledge-InstitutionsReinventing)
The other argument is that we should be seeing less embargo effect as acceptable embargoes get tightened down.
But again, how to analyse that effectively remains a challenge I don't have a good answer for at the moment
I'm going to allow myself the luxury of fifteen minutes of actual research to look at some real data, using IRL as the example
