@rosswilkinson@fosstodon.org
San Francisco, CA
Joined December 2011
- Tweets1.3K
- Following188
- Followers480
- Likes3.1K
Pinned Tweet
I sent Mum my PhD thesis; she re-wrote it as a Trump summary: “It's about bikes....... all about bikes..... oh, and people, people riding bikes. Riding bikes sitting and standing. It's very very good, you'll like it........ I like it, so.........” *widens arms and shrugs* 😂
Ross Wilkinson retweeted
Come work with me! We've got an exiting position available to join our important Integrations Team as a Product Manager. Let me know if you have any questions or want to know more linkedin.com/jobs/view/33353…
Ross Wilkinson retweeted
There is a popular idea: 'forget all other stats, just use t-tests' and I sympathize! BUT
I'd like to alter it to 'just use mixed effects models!'
so @rd_wilkinson @mel_mazzo and I wrote a paper with accompanying code to do just that
journals.lww.com/acsm-essr/A…
Ross Wilkinson retweeted
Even more new research for your weekend reading! #footwearscience
"Could a hybrid cycling-running shoe offer time savings to triathletes?"
tandfonline.com/doi/abs/10.1…
An excellent open-source textbook on improving your statistical inference by @lakens 👏
I encourage you to at least take the time to read chapter 8 on sample size justification 🤩
lakens.github.io/statistical…
To clarify, as a grad student I made every one of the mistakes I mentioned. I was lucky to have wonderful mentors. My concern is the senior researchers still making these mistakes and passing them on to students. So, I hope my comments are taken in the spirit they were intended.
(1/4)
Some comments on the quality of science at #ACSM2022. In general, aside from a few outstanding talks (mainly from comparative physiologists), the research and how it was presented was poor. The most common mistakes were: ...
(1/4)
Some comments on the quality of science at #ACSM2022. In general, aside from a few outstanding talks (mainly from comparative physiologists), the research and how it was presented was poor. The most common mistakes were: ...
(3/4)
* No description of what was considered a meaningful difference in the context of the study i.e. the smallest effect size of interest
* Describing p-values as “trending towards significance” without any reference to effect size or sample size
(4/4)
* Bland-Altman plots with a constant bias despite the clear linear trend between the error and the measured variable
Let's try and do something about it! If anyone would like to collaborate on a workshop to address these issues for #ACSM2023, please get in touch.
Ross Wilkinson retweeted
Our new study shows that data availability statements are not very useful; 1670 (93%) authors who indicated that data are available on request either did not respond or declined to share their data with us. Journal of Clinical Epidemiology: doi.org/10.1016/j.jclinepi.2…
Not often that you see a pram in the lead pack at a fun run. Well, that was @ShalayaKipp this morning. 6th place…WITH A PRAM!!! 😂 💪 💪
My annual reminder to those of us heading into conference season: remember to check your ego at the door and fight that urge to prove to others how smart you are.
Life isn’t a zero-sum game…we can all win!
Ross Wilkinson retweeted
I often see criticism on a-priori power analysis. The main criticism is we never know the true effect size. But this criticism misunderstands the goal of power analysis - you power for an effect of interest, not for the true effect! lakens.github.io/statistical…
Ross Wilkinson retweeted
Excellent topical review on 'The legacy of A. V. Hill's Nobel Prize winning work on muscle energetics' from @UQ_News 👉 buff.ly/3O0eHjY
Ross Wilkinson retweeted
Clean water any time, anywhere!
QuenchSea is the portable, handheld device that is a simple, easy to use, and turns seawater into drinkable fresh water.
A must-have, to desalinate water on the go, Pre-Order Now!
➡️quenchsea.world
#seawater #cleanwater #desalination
