Together with colleagues from Spain, we recently conducted a comprehensive simulation study examining the performance of various types of tests and methods for constructing confidence intervals for the scale coefficients in meta-analytic location-scale models. Although things are complex, generally the likelihood-ratio test and corresponding profile-likelihood intervals performed best. See here for further details: https://doi.org/10.3758/s13428-025-02622-5
#metaanalysis
Prophylactic propofol reduces emergence agitation in children after general anesthesia.
- Hoff 2015 PaediatrAnaesth #metaanalysis
→ https://metajournal.com/25917689
#statstab #319 Small Sample Size Solutions [book]
Thoughts: This should just be the default text for psychologists, as most research fits the "small sample" label.
#smallsample #book #guide #bayesian #permutation #sem #metaanalysis #nof1 #missingdata
Major Flaws in 2025 Meta-Analysis on Fluoride and Children IQ Scores
https://osf.io/preprints/osf/zhm54_v3
#HackerNews #Fluoride #Research #Children #IQ #MetaAnalysis #Health #Debate
Big thanks to Kristina Nienhaus and UniBie for a great interview about our recent #MetaAnalysis on egg hormones and fitness, published in Eco Let.
Interview ➡️ https://shorturl.at/HZHJb
📰 Article ➡️ https://doi.org/10.1111/ele.70100
💻Data & Code ➡️ https://github.com/ASanchez-Tojar/meta-analysis_egg_hormones_and_fitness
®️ Pre-registration ➡️ https://doi.org/10.17605/OSF.IO/KU47W
Ever wondered what to do with the >80k citations of the PRISMA standards for #SystematicReviews and #MetaAnalysis?
I did a bibliometric analysis of authors who use PRISMA. Join my presentation next week:
#statstab #313 Analyzing Replication Studies
Thoughts: Replications and RR are on the rise, but not all are conducted appropriately, and become meaningless.
#stats #replication #metascience #openscience #metaresearch #metaanalysis
https://lakens.github.io/statistical_inferences/17-replication.html#analyzing-replication-studies.
Any sources on handling a disproportionately larger control group (e.g. N1=150 vs. N2=4,000 - normative sample)?🙏
We set N2 = N1 to avoid overinflating the influence of the study in a meta-analysis (large N [i.e. N1+N2]), but lack references to support this approach.
#statstab #311 The analysis of continuous data from n-of-1 trials using paired cycles: a simple tutorial
Thoughts: @StephenSenn shows how to treat multiple #nof1 studies as a meta-analysis.
#sced #nof1 #metaanalysis #tutorial #clinical
https://trialsjournal.biomedcentral.com/articles/10.1186/s13063-024-07964-7
When we talk about grey literature, do we also consider grey information and grey data? I usually define it as conference papers, dissertations, and preprints only.
A worthwhile read on this topic: https://doi.org/10.1186/s13643-016-0337-y
'Open Online R Stream' (https://www.wvbauer.com/doku.php/live_streams) going live in about 20 minutes. I will talk about some of the recent updates to the metafor package.
We're delighted to present the cover of the book "Bayesian Meta-Analysis: a practical introduction", by Robert Grant and Gian Luca Di Tanna, due out on CRC Press, June 2025.
Follow this account for weekly updates with examples and tips.
Our #metaanalysis tested the adaptive significance of egg hormones in birds by combining evidence published since 1993, mostly experiments (76% effects).
In short: regardless of hormone type, we found little evidence for an effect of egg hormones on fitness-related traits, but high heterogeneity.
In 1993, H. Schwabl published in @PNASNews a seminal paper: “Yolk [as] a source of maternal testosterone for developing birds”
This was the first study proposing a link between maternal egg hormones and fitness.
Our preregistered #systematicreview & #metaanalysis in Ecology Letters synthesises 438 effects from 57 studies on 19 wild 🐦species to test if & how egg hormones relate to fitness
📰 https://doi.org/10.1111/ele.70100
Data & Code https://github.com/ASanchez-Tojar/meta-analysis_egg_hormones_and_fitness
Pre-registration https://doi.org/10.17605/OSF.IO/KU47W
Another new paper alert. We just published a paper giving a likelihood-based implementation of the Daniels & Hughes bivariate meta-analysis model for surrogate endpoint evaluations (e.g., using progression-free survival as a surrogate for overall survival): https://doi.org/10.1002/bimj.70048
We propose a bias-adjusted MLE for the variance component of the model, present simulation results, and illustrate the method with a few case studies.
1/3 In a recent paper, we described the use of location-scale models in meta-analysis to not only examine how moderators may be related to the size of the effect, but also how moderators may be related to the amount of heterogeneity among the effects: https://doi.org/10.1002/jrsm.1562
Such models can be fit with the metafor package in R: https://wviechtb.github.io/metafor/reference/rma.uni.html#location-scale-models
#statistics #MetaAnalysis #EvidenceSynthesis #ResearchSynthesis #Rstats
Although methods for meta-analyzing differences in variance between groups have been available for several years (e.g., https://doi.org/10.1111/jeb.13661), researchers still seem reluctant to adopt them widely.
Here we show an easier alternative to meta-analyze variances across species. #metaanalysis
In our latest preprint, we showcase how to use a phylogenetic meta-analysis of variance to test evolutionary hypotheses across species.
We collected data for 172 birds (314 populations) to test patterns of stabilising selection and canalisation for wing length.
#statstab #300 (!!!) 🥳🎉
Beyond the forest plot: The drapery plot
Thoughts: I can't believe they didn't call these 'tepee plots'.
I think they are cluttered, but can be useful.
#metaanalysis #pvalues #consonancecurve #pvaluefunction #dataviz #prediction #plots #figures #R
PCEA with a continuous background is associated with longer second stage and greater instrumented delivery than PCEA without a background.
- Heesen 2015 Anesth.Analg. #metaanalysis
→ https://metajournal.com/25902319
Client Info
Server: https://mastodon.social
Version: 2025.04
Repository: https://github.com/cyevgeniy/lmst