@rspfau i think if your final slide is up and names a few key points, chances are higher that folk engages with these key messages. If people dont know what to ask, the final slide could help remind them on the core messages of your talk.

🎉 Our paper on analytical flexibility in the speech sciences is finally out 🥳

Read "Multidimensional signals and analytic flexibility: Estimating degrees of freedom in human speech analyses." #OpenAccess at AMPPS:

https://journals.sagepub.com/doi/full/10.1177/25152459231162567

Nobody is surprised that twitter can‘t pay their bills anymore, right?

Twitter exodus folks, welcome to Mastodon!

This place has a few etiquette rules that may be different than what you are used to. For example, it is customary here to:

* mask potentially sensitive posts with the Content Warning flag

* use alt text on all your images

* DM your credit card number, expiration date, and security code to the author of the first toot you come across saying "welcome to Mastodon!"

Another re #Introduction for all the folks who just headed over to 🐘 after the latest bird app change:

Hi, I'm a professor for #Linguistics at the University of Oslo. I consider myself a cognitive scientist with a focus on human speech.
I'm a strong advocate for #openscience and try to help improving methodological practices across the language sciences.
Effective #scicomm is important to me and I consult and give talks on effective #dataviz and #design in research.

Final touches on the slides for tomorrow's talk on #OpenScience practices in #Linguistics! Looking forward to talking at the MultiLing research seminar @unioslo about our recent assessment with @TimoRoettger and a wonderful team of collaborators. Open to all and hybrid if anybody wants to hop on Zoom! https://www.hf.uio.no/multiling/english/news-and-events/events/guest-lectures-seminars/2023/agatabochynska.html

Want hands on training in meta-research (research on research, or the science of science)? Join our Meta-research Summer School!

When: August 28-September 1

Where: near Frankfurt, Germany

What: Participants will work with meta-scientists (including myself, Anita Bandrowski, Malcolm McLeod, Malte Elson, Maximillian Frank), in small groups, to design a meta-research study

After the Summer School: Groups may continue working to complete the project

Apply by June 16: https://www.bihealth.org/en/notices/meta-science-summer-school-designing-studies-to-improve-research-1

@adastra Das ist schwer zu vergleichen. Grade ist der NOK besonders schwach. Unabhängig davon ist die Work-Life Balance deutlich besser als zB in Deutschland. Man hat deutlich weniger admin und teaching load. Dazu kommt noch dass die Promotion zum vollen Professor sehr früh geschehen kann und mit einer substantiellen Gehaltserhöhung einher kommt.

🚨🚨🚨We are hiring

Join us at the University of Oslo as a tenured Associate Professor with a specialization in computational linguistics.

Deadline: 31st October '23

('Associate Professor' is an entry-level position. Newly minted PhD graduates are welcome to apply.)

https://www.jobbnorge.no/en/available-jobs/job/243214/associate-professor-of-linguistics-with-a-specialization-in-computational-linguistics

@bananabull I think it is problematic when you present only posterior distributions because they (a) are very narrow relative to the variance in raw data and (b) suggest very tidy distributions which just reflect the model assumptions. I think this is misleading (I have done it before but regret it)

@bananabull It's tricky. I have done it both. I think in an ideal world you present both raw data and estimates. This is doable when using the mean posterior and say a credible interval, but since raw data are usually very spread, your scale must necessarily be wide, reducing the perceived distance between conditions. Here is one way I have solved this design issue in the past by using semitransparent raw data in the background. (from https://doi.org/10.1080/23273798.2020.1853185)

🇨🇦🇿🇦🇩🇪🇨🇳🇺🇸🇬🇧
Flags are such a great design problem that if you understand the design of flags -what makes a good or bad flag- you can design almost anything...Like your slides for your next research presentation 🥼🧪

https://youtu.be/DkW4sVxOI1U

#PresentationDesign #PowerPoint

This is Part 3 of my series on presentation preparation:
Check out Part 1 on identifying your target audience: https://www.youtube.com/watch?v=BOnAPOtnHt4

And check out Part 2 on planning analog: https://www.youtube.com/watch?v=RUwjVh17aiw

To create impactful slides, you actually do not need to spend a lot of time making slides in your presentation software. Let me walk you though an efficient process for manifesting your storyboard in #powerpoint or #keynote.

https://www.youtube.com/watch?v=doNpqRk7lM8

#presentationdesign #scicomm #scientificcommunication #powerpointpresentation

One Slide to Rule them All 📽️

https://youtu.be/EG66E3ow8Gw

#scicomm #academictwitter #presentationdesign

"Academia may lose its most brilliant developers without credit models to advance their careers. Alternative, well-paid careers in industry are compelling for developers who become disillusioned in academia [...] so academic positions with transferable, in-demand skills are at risk."

Recommended short read on the need for better incentives to to reward academic software development: https://rdcu.be/c6uMN

#academia #postdoc

@EmmaHendersonRR Hey Emma, I'm using Keynote, the native mac software. But everything I did in Keynote is all doable in PowerPoint :)

Plan your presentation analog with a pen and sticky notes.

https://youtu.be/RUwjVh17aiw

Planning analog saves time in the long run and forces you to follow important design guidelines like using images, focussing content, and avoiding too much text. In this video, I will walk you through the workflow which I have been using to plan over a hundred presentations.

#design #presentationdesign #scicomm #scientificcommunication #powerpoint

@grvsmth @eagereyes @ct_bergstrom I think the bar charts are a solution. You could color the "peak" quarters and add context with text saying that the last peak was XX% of the previous one to rly drive home your message

In order to give an impactful research presentation, you must carefully consider your target audience. Your audience are the people you are giving the presentation for. The people you want to inform, to convince, or to inspire. EVERYTHING you do, you should do for them.

In this video, I discuss how to target your audience and tailor your narrative to their needs.

https://youtu.be/BOnAPOtnHt4

#presentationskills #scicomm