Ed Merkle
quantitative psychologist, open source enthusiast, recently working on R package blavaan
- Researchers often treat ordinal variables as continuous. What if we could mimic this in an ordinal factor analysis/IRT model? We propose new identification constraints so that the latent variable goes from 1 to (# of ordered categories), with connections to treating ordinal variables as continuous.
- With coauthors @winterstat.bsky.social and Ellen Fitzsimmons. Code at semtools.r-forge.r-project.org
- Consider submitting a proposal to the (open access) Psychometrika special issue on Variable Selection for Complex Psychometric Data, with a proposal deadline of Jan 15. Full details:
- Reposted by Ed MerkleA new semester is upon us! This time, I’m teaching Structural Equation Modeling. Want to follow along? Here you go: jonathantemplin.github.io/Structural-E... and www.youtube.com/playlist?lis... #StatModeling #SEM
- Reposted by Ed MerkleDon’t miss out on short courses at IMPS 2025! We have 4 short courses to choose from. www.psychometricsociety.org/imps2025-sho... #IMPS2025 #Psychometrics #QuantitativePsychology #rstats
- Reposted by Ed MerkleYou want to fit structural equation models in Julia? No problem with the StructuralEquationModels.jl package. Ships with worked examples and tutorials on how to modify the objective function easily (e.g., adding a regularization penalty). Find our preprint here osf.io/preprints/ps...
- Reposted by Ed Merkle📚😅🎉 Yay!! I just submitted the complete manuscript of my upcoming book to the publisher! Learn to easily and clearly interpret (almost) any stats model w/ R or Python. Simple ideas, consistent workflow, powerful tools, detailed case studies. Read it for free @ marginaleffects.com #RStats #PyData
- Looks like an interesting paper:
- Reposted by Ed MerkleCmdStanR v 0.9.0 released! According to Andrew Johnson, this should make Windows installation easier plus many other improvements and bug fixes discourse.mc-stan.org/t/cmdstanr-v... #Bayesian #MCMC
- Reposted by Ed MerkleHuzzah! Great news for psychometric software! Psychometrika now accepts software submissions under "Applications and Case Studies" AND software reviews in their expanded Review Section. This dual recognition elevates the status of programming in the field. #psychometrics #OpenScience
- Here is a new post providing some intuition about the proportional odds restriction in ordinal regression models. Maybe you, or your LLM, will find it useful.
- Reposted by Ed MerkleFellow probabilistic programmers I came up with a parameterization for a Cholesky factor of correlation matrices that looks good. I'm asking to test and see if it works well for you discourse.mc-stan.org/t/updated-ch... #rstats #statistics #bayesian
- Reposted by Ed Merkle{tinyplot} 0.3.0 is out! 🚨 It's a lightweight #Rstats 📦 to draw beautiful and complex plots, using an ultra-simple and concise syntax. This is a massive release! @gmcd.bsky.social @zeileis.org and I worked hard to add tons of new themes and plot types. Check it out! grantmcdermott.com/tinyplot/
- Reposted by Ed MerkleIf you know simulation based calibration checking (SBC), you will enjoy our new paper "Posterior SBC: Simulation-Based Calibration Checking Conditional on Data" with Teemu Säilynoja, @marvinschmitt.com and @paulbuerkner.com arxiv.org/abs/2502.03279 1/7
- Reposted by Ed MerkleThe Psychometric Society invites IMPS 2024 presenters to submit manuscripts to the 2024 Proceedings. Intent to Submit Deadline: February 14, 2025 Manuscript Deadline: March 21, 2025 www.psychometricsociety.org/imps2024-pro... #psychometricsociety #psychometrics #quantitativepsychology
- Reposted by Ed MerkleYou think Jacobian and Hessian matrices are prohibitively expensive to compute on your problem? Our latest preprint with @gdalle.bsky.social might change your mind! arxiv.org/abs/2501.17737 🧵1/8
- Reposted by Ed MerkleNow out in PNAS, we provide a conceptual framework for assessing replication decisions and replication reform. Do you view the scientific literature as a "book of truths" or a "book of conversations"? www.pnas.org/doi/abs/10.1... #philsci #science #SciSci #Metascience
- Reposted by Ed MerkleCall for Abstracts for IMPS 2025 The Psychometric Society Invites You to Submit an Abstract for IMPS 2025! The abstract submission portal is scheduled to open Jan 29, 2025 Please see the society website for details: www.psychometricsociety.org/post/imps-20...
- Reposted by Ed MerkleOur friend and colleague Fritz Leisch died in April last year. In a new contribution to The R Journal we honor Fritz and commemorate his many contributions to science in general and to the R community in particular. #rstats journal.r-project.org/articles/RJ-...
- Reposted by Ed MerkleI am so proud of the Psychometric Society having the courage to transition Psychometrika to Open Access. Practically all APCs are covered by TAs (see bit.ly/3NiAy6T) and for authors not covered will be waived! Support open science for everybody, send your best work to Psychometrika at bit.ly/4eXTvYp
- New preprint with @winterstat.bsky.social and Ellen Fitzsimmons on ordinal factor analysis, where the latent variables are scaled so that they generally take values from 1 to # of ordered categories. The scaling is intuitive because it is what you use when you treat ordinal variables as continuous.
- Reposted by Ed MerkleOK, here is a very rough draft of a tutorial for #Bayesian #SEM using #brms for #rstats. It needs work, polish, has a lot of questions in it, and I need to add a references section. But, I think a lot of folk will find this useful, so.... jebyrnes.github.io/bayesian_sem... (use issues for comments!)
- Reposted by Ed Merkle2. Frontiers for Young Minds Now this is an excellent initiative. If you haven't heard of this, researchers write papers targeted at kids or teenagers. But the kicker is that your target audience co-peer reviews the paper! Usually, this is done collectively as a science class
- Reposted by Ed Merkleworking extra hard on more administrative tasks so that I can clear time to properly read this - looks very interesting!! arxiv.org/abs/2412.07999 'Fast Mixing of Data Augmentation Algorithms: Bayesian Probit, Logit, and Lasso Regression' - Holden Lee, Kexin Zhang
- Reposted by Ed MerkleSuper stoked to announce that {greta} version 0.5.0 is now on CRAN! Thank you to Nick Golding for making this release happen, and for his continued support, I learnt a lot! Some features I'm really excited about: - Now uses Tensorflow 2.0 (which involved a heavy refactor of internals) - […]
- Reposted by Ed MerkleDeck.blue is good, you should check it out
- since I've gained over 1000 followers in the past hour, quick refresher on deck.blue: it's a tweetdeck-like client that I started developing back in august 2023 on the same day the old tweetdeck got paywalled i have nothing to do with twitter, the original tweetdeck team or even the bluesky team
- Reposted by Ed MerkleJorgensen & Garnier-Villarreal have a helpful paper showing several failures with "small-variance priors" in Bayesian SEM doi.org/10.1007/978-... These priors will fail for the types of misspecification we create in simulation studies, wrote a blogpost: www.jamesuanhoro.com/post/2024/11...
- Reposted by Ed MerkleI recently got to work on a fun chapter, and the preprint is now available: arxiv.org/abs/2411.04260 We provide intuitions, examples, and recommendations for running MCMC in a world where GPUs and autodiff exist (like this world!)
- Reposted by Ed MerkleHere is John Alexander Silva Díaz's latest simulation study on how SEM forest perform finding omitted influential covariates in a SEM: www.tandfonline.com/doi/full/10.... (image: CC4.0 by the original authors)
- Reposted by Ed MerkleOur Introduction to Bayesian Data Analysis for Cognitive Science (with Nicenboim and Schad) is now in production with CRC Press. It will remain freely available here: bruno.nicenboim.me/bayescogsci/
- Reposted by Ed MerkleAlong the way I translated their causal graphs into corresponding variate-covariate models (betanalpha.github.io/assets/case_...), only without forcing specific linear regression structural assumptions. In my option this makes the role of the infamous “U” variable much less mysterious.
- Reposted by Ed MerkleOur Python library BayesFlow implements methods for amortized Bayesian inference. You first train a neural network on simulated data. Then you obtain posterior inference on any real data almost instantly. Check out the dev branch for our new backend and user interface: github.com/bayesflow-or...
- I think some people on here have strong opinions about the best chalk for writing equations. Could somebody help a person who is accustomed to using whatever chalk happens to be laying around?
- Reposted by Ed MerkleThe latest `brms` CRAN release added support for `priorsense` for easy prior and likelihood sensitivity analysis doi.org/10.1007/s112... ``` > fit |> powerscale_plot_dens(variable='b_doseg', help_text=FALSE) + labs(x='Dose (g) coefficient', y=NULL) ``` #Bayesian #rstats
- Reposted by Ed Merkle🚨 New working paper about examining exam results #rstats 📈 Infer student abilities and exercise difficulties with #psychometric models ✅ Detect and interpret differential effects of exercises in subgroups of an introductory #math exam www.zeileis.org/news/exams/
- Reposted by Ed MerkleI gave a 2-hour workshop on Structural Equation Modeling and Confirmatory Factor Analysis in lavaan for the Psych #rstats Club. the recording is available here: youtu.be/YrxvV8zlNLY?... Supported by @improvingpsych.org!
- blavaan 0.5-6 was recently released to CRAN. It has always done Bayesian structural equation modeling, but in recent years we have added some functionality for ordinal observed variables and for two-level SEM (random intercepts). More info here:
- Reposted by Ed MerkleAfter a great #StanCon2024, I've started a blog series on using copulas in Stan. This first post gives a quick introduction to what copulas are. I'll keep each post short but over time I hope to put together a repository of helpful content! bggj.is/posts/stan-c... #stats
- Reposted by Ed Merklectsem 3.10.1 is on CRAN, mostly fixing a bug introduced last update when outcome variables are both continuous and binary, and including a function for improved visualisation of moderation effects. cran.r-project.org/web/packages... #psychology #rstats
- Reposted by Ed MerkleIf you want an effective color palette then don't try to get cute: just use one from colorbrewer2.org where all of the subtle work has already been done of you. If you're still tempted then read through all of kennethmoreland.com/color-advice/, and then still use ColorBrewer.
- Reposted by Ed Merklepriorsense package for prior and likelihood sensitivity analysis is now on CRAN. It works with Stan (brms, rstan or cmdstanr) fits, or just posterior draws. See more in Noa Kallioinen's post at discourse.mc-stan.org/t/priorsense... #Bayes #MCMC #rstats
- Reposted by Ed MerkleSome quantitative psychology (longitudinal emphasis) positions available with me at Uni Zurich from September on, rebleets appreciated! Postdoc: jobs.uzh.ch/offene-stell... PhD: jobs.uzh.ch/offene-stell...
- Psychometrika will convert to gold open access in 2025. The new publisher, Cambridge, has agreements with many universities (including in the US) that make the open access possible. Anyone who cannot pay the APC will be waived. www.psychometricsociety.org/post/psychom...
- Reposted by Ed MerkleLooks interesting: venpopov.com/posts/2024/i...
- Reposted by Ed MerkleI love how when we first got the internet, Wikipedia was an "unreliable source" and now it's one of the last remaining websites that's actually fact checked
- Reposted by Ed MerkleWe now have a website. udm14.com
- Reposted by Ed Merklectsem 3.10.0, for continuous and discrete time dynamic systems modelling, is up on CRAN. bug fix for certain time-varying par models; correlation approach is improved; ctACF function for auto and cross correlation plots, discussed in example here osf.io/preprints/os... github.com/cdriveraus/c...
- At last week's NCME conference, I did a workshop on Bayesian latent variable modeling with a focus on the blavaan package. We discussed factor analysis, item response models, and two-level structural equation models. Slides and code are at the first bullet:
- Reposted by Ed MerkleRead @vickiboykis.com stuff. It is good. Hat tip to @akhilrao.bsky.social
- Oh yeah! take your pick on what's relevant: On how LLMs/embeddings work starting slowly from a social network doing recommendations and going into the math: vickiboykis.com/what_are_emb... Gist of no-hype links: gist.github.com/veekaybee/be... ML in prod today: vickiboykis.com/2024/01/15/w...
- Reposted by Ed MerkleNew paper by my PhD student @matzekloft.bsky.social now published in Behavior Research Methods (open access)🥳 Measuring the variability of personality traits with interval responses: Psychometric properties of the dual-range slider response format doi.org/10.3758/s134...
- Reposted by Ed MerkleNew version of R package semtree now on CRAN: cran.r-project.org/web/packages... featuring leaner package imports.
- Reposted by Ed MerkleThis might be the best executed supply chain attack we've seen described in the open, and it's a nightmare scenario: malicious, competent, authorized upstream in a widely used library. Looks like this got caught by chance. Wonder how long it would have taken otherwise.
- Woah. Backdoor in liblzma targeting ssh servers. www.openwall.com/lists/oss-se... It has everything: malicious upstream, masterful obfuscation, detection due to performance degradation, inclusion in OpenSSH via distro patches for systemd support… Now I’m curious what it does in RSA_public_decrypt
- Reposted by Ed MerkleOur latest preprint with Ulman Lindenberger and Ethan McCormick (@emccormick.bsky.social) on "Optimal Two-Time Point Longitudinal Models for Estimating Individual-Level Change: Asymptotic Insights and Practical Implications". Enjoy at OSF: osf.io/3g2z6/
- I wrote a short post about how the R package {targets} can help your academic research project.
- Reposted by Ed Merkle🚨 #RStats 📦 developers! `altdoc` is an easy way to create highly customizable websites for your documentation. It's an alternative to `pkgdown`, with support for Quarto and several other frameworks: MkDocs, Docsify, Docute. Expand 🧵for examples. altdoc.etiennebacher.com
- Reposted by Ed MerkleSoftware bibliographies are easy! The 'softbib' package will automatically scan your project directory, detect which R packages you use, and create software bibliographies in Word, Markdown, BibTeX, and PDF formats. cran.r-project.org/web/packages...
- Reposted by Ed MerkleFREE BOOK: Psychometrics in exercises using R and RStudio by Anna Brown Great resource for university lecturers, instructors, students, and anyone who want to acquire practical skills in conducting psychometric analyses. annabrown.name/book_psychom...
- Reposted by Ed MerkleCheck out our new paper on model evaluation in the context of replication efforts. The title says it all. If you're still on twitter, I shared a long thread detailing a zine that I made to disseminate this work. I can't get threads to work in a meaningful way on bluesky, but here's the cover:
- Good fit is weak evidence of replication: Increasing rigor through prior predictive similarity checking: osf.io/n9gfa/
- Reposted by Ed MerkleA reason for the peer review crisis. Articles published grew by 50% since 2016, "which has outpaced the limited growth, if any, in the number of practising scientists. Thus, publication workload per scientist (writing, reviewing, editing) has increased dramatically." arxiv.org/abs/2309.15884
- Reposted by Ed MerkleHere is an early christmas present: Our free structural equation modeling tool Onyx is finally open source: github.com/brandmaier/o...