@DataAngler@vis.social
@DataAngler@vis.social avatar

DataAngler

@DataAngler@vis.social

I am a data scientist who programs in R and SAS. Don't ask me to pick one. I am also particularly interested in equity in education as well as gender equity and equity in policing. Occasionally fly fishing. My posts are my own.

This profile is from a federated server and may be incomplete. Browse more on the original instance.

w7voa, to random
@w7voa@journa.host avatar

Does anyone know how I can report/shut down this imposter mastodon.social account spewing spam/potential phishing?

DataAngler,
@DataAngler@vis.social avatar

@w7voa I reported them too.

DataAngler, to Birding
@DataAngler@vis.social avatar

Stunning pictures in this list!

https://www.theatlantic.com/photo/2023/09/winners-bird-photographer-year-2023/675227/

ereinbergs, to random

Hilarious opening paragraph on #r in the Croatian Medical Journal 😂 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7063554/

"Each of us has a friend with a difficult personality. However, we would not waste our time and masochistically put up with their personality if it did not benefit us in some way. And whenever we organize a get-together we always invite this friend, even though we know in advance that it would not go smoothly. It is a similar situation with R software."

DataAngler,
@DataAngler@vis.social avatar

@montenegro @ereinbergs These folks and the author of this text should do lunch

https://www.burns-stat.com/pages/Tutor/R_inferno.pdf

DataAngler, to datascience
@DataAngler@vis.social avatar

This is an excellent tutorial from @debruine on how to create a power simulation in and call the simulation n times with parameters in a dataframe. Which is so useful for wanting to check how power is affected by different aspects of study design

But I get messages in R that purrr::pmap_dfr are superseded, so are we supposed to switch to a different set of functions for passing a dataframe of parameters to a function?

DataAngler,
@DataAngler@vis.social avatar

@tanho @debruine I forgot the link: https://debruine.github.io/post/pmap_df/

MikeMahoney218, to random
@MikeMahoney218@fosstodon.org avatar

Pre-allocating vectors is for nerds https://mm218.dev/posts/2023-08-29-allocations/index.html

DataAngler,
@DataAngler@vis.social avatar

@StatGenDan @MikeMahoney218 An book set in the landscape of the Inferno makes for interesting reading.

raoulvanoosten, to statistics

I managed to simulate data and use those to calculate power for an upcoming experiment. However, power is highly variable because there is quite some variation in the real data. In some cases I need only 16 blocks for 90% power, whereas in others not even 20 blocks is enough. How would you proceed?

I used the simr package with a generalized linear model to calculate power.

DataAngler,
@DataAngler@vis.social avatar

@raoulvanoosten I think I would run the simulations with different block sizes (a thousand times each), and see which minimal block size gives you a statistically significant result at the effect size you are looking for 80% of the time.

DataAngler,
@DataAngler@vis.social avatar

@raoulvanoosten The powerSim function in {simr} lets you run the data simulation multiple times to calculate power. This resource is helpful for using simr::powerSim

https://humburg.github.io/Power-Analysis/simr_power_analysis.html

ChristophMolnar, to random

You fit a model from data.

Later you find out that your model doesn't predict new data well. Assume the new data is from the same distribution as the analyzed data.

Would you say it's valid to draw conclusions from the model?

DataAngler,
@DataAngler@vis.social avatar

@ChristophMolnar That's a good reason to be skeptical of the original model and to build a new one.

DataAngler, to HashtagGames
@DataAngler@vis.social avatar

You took the front off the car stereo and kept it in the glove compartment .

DataAngler, (edited ) to cycling
@DataAngler@vis.social avatar

"Study Will Recommend Ways to Make Chrysler Avenue Safer for Bikes"

The best headline in the history of public planning.

https://www.newarkpostonline.com/news/study-will-recommend-ways-to-make-chrysler-avenue-safer-for-bikes/article_4846370a-3bb4-11ee-bb17-438bd5e0465a.html

jhollist, to random

Hey , I suspect this is a hard one to get a firm answer on, but do we know when in October 2023 that rgdal will be archived? Yes, I waited until the last minute(s)...

DataAngler,
@DataAngler@vis.social avatar

@jhollist Blog posts about this don't seem to give an exact date...

dkraehmer, to random German

Our field experiment on code sharing behavior in the social sciences has been published in PLOS One 🥳 (co-authored by @laura_schaechtele and Andreas Schneck).

Read the full article open access https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0289380

(Main take-aways in the thread below)

#metaRep #openscience #opendata #opencode #replication

DataAngler,
@DataAngler@vis.social avatar

@dkraehmer @laura_schaechtele Hopefully we are getting to a world where it's more commonplace to push the code and shareable data into a repo that people can just use.

DataAngler, to random
@DataAngler@vis.social avatar

Grateful for the {broom.mixed} 📦 for getting tidy objects with effect estimates from linear mixed models.

DataAngler, to Birding
@DataAngler@vis.social avatar

Is this a red-shouldered hawk? My picture is blurry but I think it is one. I read that in the eastern US they can have less bright plumage.

datavisFriendly, to random

Need to generate lists 🗒️ of stimulus words for memory/learning experiments?

The WordPools package is here to help.
TWP, Paivio-etal, Battig-Montague categorized words...
https://github.com/friendly/WordPools

DataAngler,
@DataAngler@vis.social avatar

@datavisFriendly @TellLab

DataAngler, to random
@DataAngler@vis.social avatar

Get virtual backgrounds from the National Science Foundation to show your enthusiasm for the Year of

https://new.nsf.gov/public-access?utm_medium=email&utm_source=govdelivery

MedievalMideast, to ai

just changed their terms and conditions to include using anyone's video and audio for training with no option for opting out. You too can help train s!

Living with a disabled spouse, I used Zoom a lot to get through the ongoing global pandemic.

What alternatives are out there for remote teaching/meetings?

DataAngler,
@DataAngler@vis.social avatar

@MedievalMideast Yes, but paragraph 10.4 says we do consent to to use data for machine learning and AI. So which is it?

DataAngler, to random
@DataAngler@vis.social avatar

Going to present to my office on adopting research practices for reproducibility...do I dare try to learn {docker} 📦 ? It seems critical for ensuring someone else can reproduce the analyses.

DataAngler, to HashtagGames
@DataAngler@vis.social avatar

Cormorant Song

stevensanderson, to opensource
@stevensanderson@mstdn.social avatar

Group percentages in R with and
#R

https://www.spsanderson.com/steveondata/posts/2023-07-24/

DataAngler,
@DataAngler@vis.social avatar

@stevensanderson I have a function for this because it is so common a task.

t_Likert = function(df,var) {
df |>
select({{var}}) |>
drop_na() |>
group_by({{var}}) |>
summarise(Freq = n()) |>
mutate(Pct = round(Freq/sum(Freq),3))
}

TexasObserver, to Virginia
@TexasObserver@texasobserver.social avatar

Earlier: Campaign finance laws in and contain loopholes big enough to drive a truck through. Or, in this case, a million dollar donation from a sketchy shell company.

https://www.texasobserver.org/joe-liemandt-glenn-youngkin-donation-austin/

DataAngler,
@DataAngler@vis.social avatar

@TexasObserver And of course the company registered in Delaware, where anonymous, shady dealings are allowed to thrive.

DataAngler, to random
@DataAngler@vis.social avatar

I’m now officially obsessed with using narrow fonts like Roboto Condensed in all my charts. It’s readable and a space saver.

DataAngler, to random
@DataAngler@vis.social avatar

Megan Rapinoe announces her retirement following the 2023 NWSL season

https://www.washingtonpost.com/sports/2023/07/08/megan-rapinoe-announces-retirement/

Private
DataAngler,
@DataAngler@vis.social avatar

@dougholton @edutooters 'Twas every thus. 🙂

https://journals.sagepub.com/doi/full/10.3102/003465430298487

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • JUstTest
  • rosin
  • thenastyranch
  • ethstaker
  • osvaldo12
  • mdbf
  • DreamBathrooms
  • InstantRegret
  • magazineikmin
  • Youngstown
  • khanakhh
  • Durango
  • slotface
  • kavyap
  • provamag3
  • cubers
  • everett
  • cisconetworking
  • tacticalgear
  • ngwrru68w68
  • Leos
  • GTA5RPClips
  • normalnudes
  • modclub
  • tester
  • anitta
  • megavids
  • lostlight
  • All magazines
    •