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. 😀

#statstodon #statschat
In the absence of a Registered Report how would one publish, say, a study on an improved measurement if you find in a Bland-Altman plot (Turkey mean-difference) the error/bias is too big? Namely, your "new" measure isn't good enough.

What would you do?

#statstodon #statschat
In the absence of a Registered Report how would one publish, say, a study on an improved measurement if you find in a Bland-Altman plot (Turkey mean-difference) the error/bias is too big? Namely, your "new" measure isn't good enough.

What would you do?

what ideas to improve #Statistics would you present a group of #psychology master students?
On my list so far are:

1. #Bayesian statistics (plus critique of frequentist)

2. use simulations to understand #data generation processes + use it for power analyses (e.g. #SuperPoweR from @lakens )

3. think about whether you want to test a hypothesis or estimate an effect & how to design experiments accordingly (inspired by @EJWagenmakers )

4. maybe #causal #analysis

#OpenScience #StatsTodon

People interested in this thread might enjoy the "Data Never Speak for Themselves" section from Nancy Krieger's article:

"Structural Racism, Health Inequities, and the Two-Edged Sword of Data: Structural Problems Require Structural Solutions"

https://www.frontiersin.org/articles/10.3389/fpubh.2021.655447/full

#Epidemiology #EpiVerse #DataScience #Science #Statistics #Statstodon

I believe https://bayes.club is now on Mastodon v4, so you should be able to follow hashtags:

#julialang
#bayes
#bayesian
#stats
#statstodon

@atezcan look for the hashtags of your favorite topics and pin them to a feed (or just do a search) then follow all the people 😂

#epidemiology #epiVerse #epitwitter
#biostatistics #biostats #statstodon #statstwitter #rstats etc.

The reflex to indiscriminately stratify and adjust on covariates is strong in biomedical research. I face this so often, especially with clinician-scientist collaborators.

The analyses and models that arise from this, in fact, often do not well represent how exposures and treatments exert their effect in reality.

A #TargetTrial approach crystallizes when, where, and how co-variates should be treated.

Take this great illustration from Target Trial originator @MiguelHernan where he explains the study design choices for study on COVID vaccine booster effectiveness:

https://fediscience.org/@MiguelHernan/109331025655365201

#epidemiology #biostatistics #epitwitter #statstodon #epiVerse @epiVerse

@alexpghayes @paulgp @jhaluska #statstodon, statstodon.social, stats.social etc?

Does there exist...

... #mathstodon ?
... #statstodon ?

... #ml_todon ?
... #dl_todon ?
... #rl_todon ?

... #tech_todon ?
... #sci_todon ?