Post-doc in #Psychology and #CognitiveNeuroscience at #unieichstaett studying #PerceptualDecisionMaking, #VisualAwareness, and #Metacognition using #CognitiveModeling, #Psychophysics and sometimes #EEG. Also interested in #rstats, #python, #openscience and #MetaPsychology.
I looked at 325 psychology study replication results today, and the best simple rule I found for telling if a study would replicate is this:
When the original study's p-value was at most 0.01, about 72% of the papers replicated.
Whereas when p>0.01, only 48% replicated.
Others have found similar results.
For instance, @uebernerd has suggested p=0.005 as a cutoff:
The {purrr} package is a powerfull way to replace loops. The {furrr} package takes this approach one step further by parallel execution: https://furrr.futureverse.org/index.html #rstats #parallel
Now out in Psychological Medicine: Our paper demonstrating a ‘leapfrog’ trial as a method to accelerate the development and optimization of psychological interventions:
💰https://www.cambridge.org/core/journals/psychological-medicine/article/abs/demonstration-of-a-leapfrog-randomized-controlled-trial-as-a-method-to-accelerate-the-development-and-optimization-of-psychological-interventions/55D66C3E24BC026E92927256EA0D3443
:cc: https://psyarxiv.com/dvzn4/
#OpenScience: #Preregistered, with #openData and #openMaterial.
(Posts taken from @Blackwell_SE@twitter.com)
1/
The Bayesian One-sample t-Test Supersedes Correlation Analysis as a Test of Validity https://share.osf.io/preprint/E0200-826-CFF
@nicebread
Gronau et al. https://doi.org/10.1037/xge0000324 estimate that 40% of all significant p-values published in Psychonomic Bulletin & Review stem from the H0. This is not want because it's conditioned on being significant and published. 60% could be regarded as upper limit for p(H1), though.
Bad news: Collecting data on #MTurk may essentially boil down to asking a crowd of #bots for their opinion:
Too Good to Be True: Bots and Bad Data From Mechanical Turk
(by Margaret A. Webb & June P. Tangney)
https://journals.sagepub.com/doi/abs/10.1177/17456916221120027
Quotes:
"at best, I had gathered valid data from 14 human beings—2.6% of my participant sample (N = 529)"
"This article is not meant as an empirical assessment of the validity of all MTurk data; rather, it is an illustration of an individual experience. "
#Introduction
Hi there #AcademicMastodon! I'm Manuel, a postdoc in #Psychology and #CognitiveNeuroscience at #unieichstaett. I study #PerceptualDecisionMaking, #VisualAwareness and #Metacognition using #CognitiveModeling, #Psychophysics and sometimes #EEG. I am also interested in #rstats, #python, and #openscience.
Vacancy in our AI Department! -- Assistant Professor of Cognitive Robotics at the Donders Centre for Cognition https://www.ru.nl/werken-bij/vacatures/assistant-professor-of-cognitive-robotics
Added a bunch of custom program language emojis!
:rstats: -- : rstats :
:stata: -- : stata :
:python: -- : python :
:julia: -- : julia :
I've made a bot that toots preprints from PsyArXiv: https://botsin.space/@PsyArXivBot I liked it on the birdapp, and I missed it here. Hope it is useful for people.
It is fast, but the hyperlink goes to the SHARE meta-data. You need an extra click to go to the actual preprint. If you know how to fix this (I am using IFTTT) let me know.
