mastodon 4.8.0-nightly.2026-10-06Lecturer Psych&Crim @KingsCollegeLon | Deception Detection; Emotions; JDM | Open Science; R; Bayes | @ukrepro ReproTea & StatsTea | #statstab | 🇷🇴 🇬🇧🌍
#statstab #525 What Good Is a Regression? Inference to the Best Explanation and the Practice of Political Science Research
Thoughts: It's interesting to see how other fields consider the problem of knowledge generation
#524 Interactions in Logit Regressions: Why Positive May Mean Negative
Thoughts: Nonlinear models require a bit more caution for new researchers when making inferences.
#statstab #522 Bayes Rules! Different priors, different posteriors
Thoughts: Nice illustration of how uninformative and informative priors change your posterior.
#statstab #520 Reverse‐Bayes methods for evidence assessment and research synthesis
Thoughts: I was reminded of this paper on assessing the evidentiary value of a finding. What do ppl think?
#bayes #inference #evidence #probability #priors #sensitivity
#statstab #519 Sensitivity Analysis and Robustness Checks
Thoughts: Multiverse analysis allows you to capture decision uncertainty.
#multiverse #jackknife #bias #robustness #sensitivity #r #outliers
https://mike-data-analysis.share.connect.posit.cloud/sensitivity-analysis-and-robustness-checks.html
#statstab #524 {simglm}: Tidy simulation and power analyses
Thoughts: As your design becomes more complex, simulation is the only way to go.
#simulation #r #rstats #tidy #tidyverse #poweranalysis #design #tutorial
#statstab #531 Using Extant Data to Improve Estimation of the Standardized Mean Difference
Thoughts: I wonder if it's a good idea. Any edu/clinical people to chime in? Seems like using historic data.
#statstab #530 Fisher vs Stouffer
Thoughts: Combining p-values (say in a meta-analysis) is not a straightforward thing. Here are two methods.
#fisher #stouffer #pvalues #metaanalysis #zscores #evidence #falsepositive
#r #simulation
#statstab #529 Understanding and misunderstanding randomized controlled trials
Thoughts: Randomization is close to magic, but we still must be critical of our methods.
#randomization #rct #design #critique #covariates #causalinference #ATE #confounder #bias
#statstab #528 Survey Bot Detection Workshop
Thoughts: "Pick a cutoff value for detecting random responders" can you spot the bots?
#OpenScience #bots #detection #sample #human #participants #deception #shinyapp #education #tutorial
#statstab #526 A Meta-Analysis of the Impact and Heterogeneity of Explicit Demand Characteristics
Thoughts: The impact of the design on conclusions is under-appreciated
#demandcharacteristics #design #methodology #error #bias
#metascience #metaanalysis
#statstab #523 Equal vs. Proportional Weighting in Least-Squares Means
Thoughts: Interesting comparisons of bias for the two methods. Complex stuff.
I wonder what people make of this.
#weighting #error #bias #LSM #emmeans #covariates
https://imazubi.github.io/stats/stats_equal_vs_proportional_lsm/index_equal_vs_proportional_lsm.html
🚨 Come to iIIRG 2026 to hear our talk about better methods to analysing interview data! W/ Dr Cody Porter and @mattansb@scicomm.xyz
#statstab #521 Non-adjustment for multiple testing in multi-arm trials of distinct treatments: Rationale and justification
Thoughts: Most researchers struggle with multiplicity corrections.
