#statstodon — Public Fediverse posts

If I were to say that the primary benefit of randomization (and possibly blinding) is that it makes simple statistical models match the true data generating process quite well, would that be a provocative statement? Or would that be obvious? Is there a good reference for this line of reasoning?

#stats #statstodon #phil #philosophy #philsci #philscidon

Great guide to thinking about the age-period-cohort problem by Julia Rohrer: https://osf.io/preprints/psyarxiv/8zmuv

#stats #statstodon

In the latest installment of "Statistical Ideas that Changed the World", son Ryan interviews father Rob Tibshirani, humor included.
https://www.youtube.com/watch?v=3rvl4KV41JE&list=PLt_pNkbycxqahVksaNnjz3M6759xHIZ-r&index=15
#RStats #statstodon #machineLearning

@vbuendiar I used Statistics for the Life Sciences by Myra Samuels when I was at uni and I've consulted it occasionally since. One of these open access intros to statistics may suit you:
https://openstax.org/details/books/introductory-statistics
https://open.umn.edu/opentextbooks/textbooks/459
https://openintro-ims.netlify.app
https://directory.doabooks.org/handle/20.500.12854/34734
#Statistics #StatisticsBasics #StatsEducation #StatsTeaching #Statstodon

Hello #statstodon! A reviewer asks me to perform a post-hoc #PowerAnalysis. I know this is generally not advised because if you replace the a priori effect size by the effect size measured in the experiment, this will introduce an erroneous relationship between the significance level of the test and the measured power.
… but does that mean that there is no proper way of measuring power retrospectively? For example, if you refrain from using the measured effect size and instead simulate a range of “a priori” effect size unrelated to the results of the test, then the dependency of the power to the significance level should not happen?
#stats #statschat @lakens

I have long-term issues getting planned missingness right.
Always seem to not be enough for the sample size for FIML or multiple imputation.

End up with missing covariance elements & nasty error messages.

Any good resources on figuring out %missing & N for planned missingness?

#stats #statstodon

@philipncohen @WeedenKim

It seems like every picture from the strike has multimillionaire in the center?

I mean, the distribution of compensation between actors and movie&tv executives seems very similar. Top 10 actors get the same as top ten movie executives (~ $400 million) and then it starts to decrease but familiar actors still get the same as execs just below C-suite.

#SAG #statstodon

@DToher

#Statstodon is apparently a thing. They might want in on that. 😀