Lightnews — Scholar-powered news

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 1h

😂

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 2h

Nice plots! BTW, would you consider expanding the section on measures of bias? I've found these to be hardest to parse when reading old articles as dif authors use dif symbols and terms for the same thing (or vice versa). Even a sup. doc with the different ones and R code would be super useful.

1 1

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 1d

When do you guys just stop a model? I'm feeling a lot of sunken cost here and want to keep going...

(first time running a Dirichlet family model)

2

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 1d

This is still running BTW...

1 1

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 2d

Also, most theories do not have that level of complexity to permit adding moderators so easily. At least not in psych.

1

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 2d

I'm not surprised to see anything in pnas (iykyk) :p

Adding a moderator shifts you from ATE to a CATE. That may not be your original question. If it is, then fine.

But, adding lots of moderators increases the prob that at least one sub-group shows *something*, even if spurious.

2 1

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 2d

This is what I find awesome (in the truest sense) about scientific development in the modern age. One group posts a new technique and within a few days it is implemented by others. #rstats is magic (and efficient)

Magnus Johansson @pgmj.bsky.social · 2d

This is really neat. I have borrowed the reliability() function to my `easyRasch` package, and use plausible values instead of fully Bayesian estimation to produce similar estimates/CIs, see code example below. RMU point estimates are similar to EAP reliability.

pgmj.github.io/easyRasch/re...

Giacomo Bignardi @bignardi.bsky.social · 12d

New preprint with @rogierk.bsky.social @paulbuerkner.com - we introduce "relative measurement uncertainty" - a reliability estimation method that's applicable across a broad class of Bayesian measurement models (e.g., generative-, computational- and item response theory-models osf.io/h54k8

1 7

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 2d

Is it the reliability of per person d-prime estimate? (I am out of cognitive energy for the week)

1

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 2d

Would these be useful for our SDT model? Could it be used to say something new?

1

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 2d

🤣🤣🤣

1

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 2d

#statstab #435 Guide to understanding the intuition behind the Dirichlet distribution

Thoughts: Useful for composite proportions, but take ages in brms.

#brms #Dirichlet #proportions #modelling #rstats #r #betareg

www.andrewheiss.com/blog/2023/09...

Guide to understanding the intuition behind the Dirichlet distribution | Andrew Heiss

Learn about the Dirichlet distribution and explore how it’s just a fancier version of the Beta distribution

www.andrewheiss.com

3

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 3d

R: How many cores do you want to use?
Me: Yes.

#R #brms #bayesian #rstats

3 11

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 3d

To be more precise (although I haven't spent much on the paper): heterogeneity and confounding aren't the same. Moderators help with the former. However they also change your Estimand. I don't like that as a solution.

1

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 3d

If you are an experimental researcher and can randomize the this is unnecessary. Most often the Obs study ppl and the experimental ppl don't talk to each other.

1 1

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 3d

#statstab #434 Exposing omitted moderators: Explaining why effect sizes differ in the social sciences

Thoughts: Maybe our models are too simple to makdle the generalisable claims we want.

#bias #moderator #causalinference #heterogeneity #effect

doi.org/10.1073/pnas...

Exposing omitted moderators: Explaining why effect sizes differ in the social sciences | PNAS

Policymakers increasingly rely on behavioral science in response to global challenges, such as climate change or global health crises. But applicat...

doi.org

2 3 3

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 4d

#statstab #433 The Table 2 Fallacy: Presenting and Interpreting Confounder and Modifier Coefficients

Thoughts: Adding predictors to your model results in marked changes in inference. Be mindful!

#table2 #fallacy #covariates #bias #inference #regression

doi.org/10.1093/aje/...

The Table 2 Fallacy: Presenting and Interpreting Confounder and Modifier Coefficients

Abstract. It is common to present multiple adjusted effect estimates from a single model in a single table. For example, a table might show odds ratios for

doi.org

4 4

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 5d

🤣

1

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 5d

Indeed. Diamond open access is more appropriate

1

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 5d

#statstab #432 PsychOpen Gold

Thoughts: instead of submitting to greedy and unhelpful publishers, try this list of fully open and free journals in psychology.

#OpenScience #openaccess #apcs #goldaccess #pedagogy

psychopen.eu

PsychOpen GOLD: Open Access Publishing

We are a Diamond Open Access platform for psychology research. Peer-reviewed, free-to-read journals with no publication fees, promoting open science.

psychopen.eu

1 2 9

Reposted by Dr Mircea Zloteanu ☀️ 🌊🌴

Matti Vuorre @matti.vuorre.com · 16d

100% this, especially with Wiley given their anti-preprint stance.

Francesco Poli @francescopoli.bsky.social · 17d

Why are we still sending our work to Wiley and other publishing companies so that they can profit from it? There's so many better options now, for example: psychopen.eu/journals/

1 5 14

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 6d

My condolences.

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 6d

PowerPoint is still king for posters!

Reposted by Dr Mircea Zloteanu ☀️ 🌊🌴

Ruben C. Arslan @ruben.the100.ci · 6d

We're looking for an experimental psychologist in Witten. Contact me if you have questions about the job (I'm not on the commission, so can talk freely :-)
uni-wh.softgarden.io/job/59638545...

Universitätsprofessur für Experimentelle Psychologie (W2-analog) bei Universität Witten/Herdecke

Wir suchen: Universitätsprofessur für Experimentelle Psychologie (W2-analog) (Vollzeit) • Alfred-Herrhausen-Straße 50, 58455 Witten, Deutschland

uni-wh.softgarden.io

14 12

Dr Mircea Zloteanu ☀️ 🌊🌴 @mzloteanu.bsky.social · 6d

#statstab #431 Attention Check Items and Instructions in Online Surveys with Incentivized and Non-Incentivized Samples: Boon or Bane for Data Quality?

Thoughts: We need more discussions on the "bad respondents" issue in research.

#Survey #quality

papers.ssrn.com/sol3/papers....

Attention Check Items and Instructions in Online Surveys with Incentivized and Non-Incentivized Samples: Boon or Bane for Data Quality?

In this paper, we examine rates of careless responding and reactions to detection methods (i.e., attention check items and instructions) in an experimental sett

papers.ssrn.com

1

Reposted by Dr Mircea Zloteanu ☀️ 🌊🌴

Richard D. Morey @richarddmorey.bsky.social · 9d

"Ultimately, contemporary AI is research misconduct"

I like it already :)

1 3 16