Lightnews — Scholar-powered news

Robin Blythe @rbly.bsky.social · 12h

I think the issue is that a lot of self-cites are gratuitous, often coming from senior faculty who have systematically inflated their metrics with cites that are barely relevant. However, totally agree with the message that you should be building towards a nuanced and complete understanding.

3

Robin Blythe @rbly.bsky.social · 14h

I see this a lot and the reason (IMO) is that it's a simple way to end the paper without deeply engaging in either the mechanism for change or the implementation factors that lead to the program falling apart as soon as funding goes. It's often a copout

1

Robin Blythe @rbly.bsky.social · 1d

You might enjoy the Raven's Mark series by Ed McDonald, one of my favourites. Basic concept is that the world is a magically corrupted hellscape due to the squabbles of demigods, each of whom maintains a city-state as their principle place of worship. It's great!

2

Robin Blythe @rbly.bsky.social · 2d

Would you call this a dual Y? More so just two labelled points I think

Robin Blythe @rbly.bsky.social · 2d

Registry data means we couldn't really adjust for confounding, but we think our results are potentially understated. RCTs have shown lower complication rates for cemented hips, despite more frail adults receiving cemented fixation. We need longitudinal outcomes data to investigate this finding.

2

Robin Blythe @rbly.bsky.social · 2d

New paper in Journal of #Arthroplasty estimating lower costs and QALY gains from ensuring US #Medicare hip fracture patients receive cement for femoral stem fixation. Big thanks to the American Joint Replacement Registry for enabling this study through sharing data:
doi.org/10.1016/j.ar...

Redirecting

doi.org

1 2

Robin Blythe @rbly.bsky.social · 3d

The relationship between epi and stats is a strange one as, in my experience, the tools are often the same but the assumptions and meaning very different. Causal inference and stats practitioners appear to frequently butt heads.

2 2

Robin Blythe @rbly.bsky.social · 4d

I'll set up a sharable link after journal formatting!

1 2

Robin Blythe @rbly.bsky.social · 4d

New paper! I'll go into a bit more depth about it when I'm back from a work trip, but this was a fun study to conduct using #rstats and the peer review process was really excellent. #statsky

Journal of Clinical Epidemiology @jclinepi.bsky.social · 4d

Continuous predicted risks should be retained when deploying clinical prediction models - Journal of Clinical Epidemiology www.jclinepi.com/article/S089...

Continuous predicted risks should be retained when deploying clinical prediction models

Clinical prediction models are used to obtain predicted risks of a diagnosis or future event. While models can produce continuous predicted probabilities, these are often dichotomised or categorised i...

www.jclinepi.com

1 5

Robin Blythe @rbly.bsky.social · 5d

I've heard that there's sort of a "two schools" schism emerging in NLP, seemingly worse than what's been happening in the stats discourse. Guaranteed there are plenty of researchers running sensitive info through OpenAI servers

1

Robin Blythe @rbly.bsky.social · 5d

Dealing with terminology creep in healthcare is infuriating because docs throw around "AI" and they usually just mean a random forest but can't differentiate it from LLMs or even a logistic regression

1 2

Robin Blythe @rbly.bsky.social · 5d

The sad thing for me is the total dismissal of the foundational theory required to understand what is appropriate, why it works, and subtle differences in methods and applications that lead to huge differences in meaning. Coding is just the pen, without theory there is no meaning.

1 1 5

Robin Blythe @rbly.bsky.social · 5d

There's a good interview with Romney out there where he says people like Ron Johnson are lovably credulous about everything but he has no respect for liars like JD Vance

1

Robin Blythe @rbly.bsky.social · 6d

I'm sure there are a lot of pins like the National Lard that we are all too mature to make.

2

Reposted by Robin Blythe

Mattan S. Ben-Shachar @mattansb.msbstats.info · 6d

There was an excellent tweet years ago that I can't find, someone announced their departure:

"I'm leaving academia to do the things I love most: research and teaching"

2 11

Robin Blythe @rbly.bsky.social · 6d

Lol there was absolutely a prodigious quantity of tomato juice being imbibed

1

Robin Blythe @rbly.bsky.social · 6d

I think this is a universal phenomenon. Can't remember the exact post, but this guy discusses it a lot in an Australian/tech context: ludic.mataroa.blog
(All his posts are great!)

Ludicity

<blockquote> "While I'm deeply sympathetic, the author should be discussing their issues with a therapist rather than spreading this on the internet." </blockquote> I finally rage-quit and started a...

ludic.mataroa.blog

1

Robin Blythe @rbly.bsky.social · 6d

Now that you mention it, I assumed those were Americans 🤔

1 1

Robin Blythe @rbly.bsky.social · 6d

You may be right, but the higher ups in my dept still spend a good 50% of their time (at least) doing the management part. Better than 100% I guess!

1

Robin Blythe @rbly.bsky.social · 6d

Reading this makes me think I'm maybe just lucky - I have enough free (work) time to pursue research I consider meaningful, and my funded research supports govt policymaking. What Mike describes is the black hole of grants and management I'm trying to avoid long term. Not sure how yet, though.

1 2

Robin Blythe @rbly.bsky.social · 6d

I am transiting SG-NY via Frankfurt. All older Germans on my flight are dressed as though they are prepared to scale Kilimanjaro at a moment's notice. Audible complaints were heard when the only remaining breakfast was spicy seafood noodles.

2 5

Robin Blythe @rbly.bsky.social · 7d

You can understand plenty about the mechanisms of the citric acid cycle. Still got to memorize a huge amount of shit though lol

1 5

Robin Blythe @rbly.bsky.social · 7d

Greatest con the fossil fuel companies ever pulled was by telling everyone that emissions are just down to us.

1

Robin Blythe @rbly.bsky.social · 7d

Agreed. Does my head in that people think AI is going to change all that somehow, completely ignoring everything we've learned in the last decade(s)

1

Robin Blythe @rbly.bsky.social · 7d

Lmao this would have been priceless to see