easystats
Official channel of {easystats}, a collection of #rstats 📦s with a unifying and consistent framework for statistical modeling, visualization, and reporting.
“Statistics are like sausages. It’s better not to see them being made, unless you use easystats.”
- [Not loaded yet]
-
View full threadOne advantage of that data is that it has labelled data, and you can see the automatic labelling feature in later plots.
- Ah, no, we used `datawizard::to_factor()` to convert label attributes into factor levels.
- 🎉 Great news for #rstats users! If you love the native R graphics feel of #tinyplot AND you're a fan of the powerful #easystats #modelbased package, this is for you! Thanks to @gmcd.bsky.social, we significantly enhanced the tinyplot integration. 🔗 Read more: easystats.github.io/modelbased/a...
- Alrighty, {easystats} users! 👋 Ever wonder how those neat tables magically appear in your R console, or even better, in your fancy #rstats Markdown and Quarto docs? Well, most of the objects you work with in {easystats} are basically tables, i.e. a 2D matrix with columns and rows...
-
View full threadNot sure about the specific requirements for APA 7 style, but I guess you may need some additionally tweaking of the returned table object.
- ... which you can do by adding additional "layers", if you use the gt-format or tinytable-format.
- That "tt" option is now fully rolled out across several #easystats packages, powered by the amazing {tinytable} package. This means you can create tables in a gazillion different output formats! How cool is that? 🤯
- Wanna dive deeper into the table universe? Check out these links: 👉 easystats.github.io/insight/arti... 👉 vincentarelbundock.github.io/tinytable/ Happy printing, everyone! 🖨️ #rstats #easystats
- ... and when they print, it's thanks to some behind-the-scenes magic with `insight::format_table()` and `insight::export_table()`! ✨ But there's more! Many #easystats functions also have a `display()` method. Think of it as your personal table stylist, making everything look super user-friendly! 💅
- And you can totally control the vibe! Use the `format` argument to get "markdown" (for a classic kable look), "html" (for a sleek gt-table), or the new kid on the block, "tt" (for a tinytable masterpiece!).
- Okay, so you've crunched your numbers and got some awesome statistical models? Sometimes, just knowing "X predicts Y" isn't enough to really get to the juicy bits. That's where the cool post-hoc stuff comes in – think estimated marginal means, contrasts, pairwise comparisons, or #marginaleffects.
-
View full threadDealing with interrupted time series where a sudden event just messed with everything? easystats.github.io/modelbased/a... Curious about disparities, different trajectories of hidden groups, and what makes them tick? easystats.github.io/modelbased/a...
- Even if you're not tackling these super complex questions, {modelbased} is generally just a fantastic tool for really getting your head around your statistical models. Go on, take a peek! You might just fall in love: easystats.github.io/modelbased/ #rstats #easystats #marginaleffects #inference
- True to the #easystats vibe, {modelbased} keeps things simple, flexible, and easy-peasy so you can truly unleash the power of your models without pulling your hair out. Ever wondered about cause and effect in observational data without needing a time machine? easystats.github.io/modelbased/a...
- Got a thing for social and health inequalities? easystats.github.io/modelbased/a... Or maybe you're into the nitty-gritty of intersectional analysis? easystats.github.io/modelbased/a...
- The {modelbased} R package is here to be your statistical sidekick! It's an #rstats gem that helps you squeeze every last drop of insight from your models. It's got a super user-friendly interface to pull out all those estimands from a huge variety of models (doi.org/10.21105/jos...).
- How to summarize the total effect of a categorical variable like education? A new vignette shows how to compute absolute and relative inequality with the #easystats {modelbased}📦in #rstats. Get a single, interpretable number to quantify overall group disparities! easystats.github.io/modelbased/a...
- 🎉 Great news, R users! 🎉 We're thrilled to announce that {tinyplot} support is coming to the #rstats #easystats project! Get ready for even more amazing stuff to make your data analysis a breeze! 📊✨ @gmcd.bsky.social @vincentab.bsky.social @zeileis.org
- As you can see, many plot types already work, just some fine-tuning left to do...
- Just dodging is not yet implemented in {tinyplot}, but hopefully coming soon!
- Improved support for the great {tinytable}📦 from @vincentab.bsky.social coming to the easystats packages! Use the `display()` method for different output formats of your tables - HTML, markdown, or - when `format = "tt"` a `tinytable` object that renders context-dependent. #easystats #rstats
- Here's the default HTML rendering.
- Since `display(format = "tt")` returns a `tinytable` object, you can easily modify the table to meet your needs.
- bayestestR::describe_posterior() works on rvar columns
- [Not loaded yet]
- But I think the rvar-support is more recent ;-)
- [Not loaded yet]
- [Not loaded yet]
- You may find some of the resources useful, e.g. there are two sets of slides for the {modelbased} package: easystats.github.io/easystats/ar... (and of course the website, which gets regular updates: easystats.github.io/modelbased/)
- Several easystats📦were updated the past weeks, make sure to install them to get the latest features! Here's what's new: - 📦insight, bayestestR: performance improvements for Bayesian models, better support for brms-mixture models 1/2 #easystats #rstats easystats.github.io/easystats/
- - 📦parameters, performance: improvement to all functions related to factor or principal component analysis, as well as psychometric testing (Cronbach's alpha, omega...) - 📦modelbased: better support for brms-mixture models More updates in other packages are in the pipeline...
- Yay, we have reached the 30 million downloads mark (and > 10k citations of our packages)! #easystats #rstats (nice metrics, despite not 100% accurate, but still...)
- Since we got questions regarding if model predictors also predict class membership or only the mean outcome for each class, we added a short paragraph including a summary table and some example code at the end of the vignette, clarifying the different GMM options: easystats.github.io/modelbased/a...
- Unlock hidden patterns in longitudinal data! 🚀 Our new vignette shows how to use brms & easystats to perform Growth Mixture Models, identify unique developmental trajectories, and visualize & interpret your findings with ease. #rstats #brms #easystats easystats.github.io/modelbased/a...
- Unlock hidden patterns in longitudinal data! 🚀 Our new vignette shows how to use brms & easystats to perform Growth Mixture Models, identify unique developmental trajectories, and visualize & interpret your findings with ease. #rstats #brms #easystats easystats.github.io/modelbased/a...
- Note that these new features require the current GitHub versions of our packages, which you can install via `easystats::install_latest()`.
- Shout out to the #rstats {performance} package. Its check_model() function is super helpful for model evaluation. #statistics #STEM
- One annoyance worth mentioning: @easystats.bsky.social, the VIF values are often extremely high when interactions are present. Why not use the same approach as the vif() in the {car} package? I also like that the car package presents generalized VIFs when appropriate.
- We do use the same approach as car::vif() (we actually report gvif). If you have an example where our output deviates from car::vif() please do share!
- We're happy to have an accompanying publication for another #rstats #easystats package published! Thanks to @vincentab.bsky.social and @tjmahr.com for reviewing the manuscript!
- Just published in JOSS: 'modelbased: An R package to make the most out of your statistical models through marginal means, marginal effects, and model predictions' doi.org/10.21105/joss.07969
- [Not loaded yet]
- And there's a non-documented option "specific", which is equivalent to `estimate_relation()`.
- Page 3 in the paper (or the docs: easystats.github.io/modelbased/r...)
- [Not loaded yet]
- Yes, see argument `estimate`. Using this argument, you should be able to easily reproduce results from emmeans and marginaleffects::avg_predictions. It's more a matter of naming things/wording, where the modelbased approach differs from emmeans or marginaleffects.
- 🆕 Introducing check_group_variation() in the {performance} #Rstats package! 🎉 This function makes it easy to checks if variables vary within or between levels of grouping variables. Perfect for understanding and designing mixed models 🚀 easystats.github.io/performance/... #stats #easystats
[Not loaded yet]- Messy data is our bread and butter!
- [Not loaded yet]
- Yes, exactly 💯
- This function can also be used to detect any predictors that might cause heterogeneity bias - variable that vary both within and between groups, that can be treated with datawizard::demean() easystats.github.io/datawizard/r...
- One function per week, this time we look closer at random effects variances in mixed models: `performance_reliability()` & `performance_dvour()`. Is the variability in your data due to noise within groups, or actual differences between groups? #easystats #rstats easystats.github.io/performance/...
- [Not loaded yet]
- It's slightly different, it adjusts for trial count.
- In case you missed it, we recently updated some of our packages, including many new features (again) in the #rstats #easystats {modelbased} package: easystats.github.io/modelbased/n... The last weeks we were working a lot on improving support and performance for Bayesian models and especially
- improved our support for the flexible {brms} package (like better handling of the many possible distributional parameters). Mixed effects models and predicting or testing their random effects has also been on our list, both for Bayesian and frequentist models.
- We are also implementing features for data/designs that are typical in the field of psychology, e.g. reaction times / decision making. E.g., support for Drift Diffusion models was added recently. More to come!
- One function per week (maybe we change it to month?), this time showing how to easily create a table of a sample description using the #rstats #easystats {report} package: easystats.github.io/report/refer... Appropriate summary automatically applied based on variable types, also supports weighting.
- Explore the many ways to interpret statistical models in #rstats using the #easystats {modelbased} package. There is a series of five vignettes, demonstrating how to easily answer different research questions. No longer struggle with confusing coefficient tables! easystats.github.io/modelbased/a...
- For every (possibly) complex question, there's an clear and easy to communicate solution - if you go for predictions/marginal effects/(pairwise) comparisons/contrasts instead of trying to interpret coefficients. #easystats #rstats
- [Not loaded yet]
- [Not loaded yet]
- You can also live risky and install the latest development-versions, where binaries are provided from r-universe: `easystats::install_latest()`. If there are any issues, you can revert back easily using `easystats::install_latest("cran")`.
- A new version of {modelbased} just hit CRAN, including bug fixes and many new features. modelbased let's you easily compute marginal means, contrasts and pairwise comparisons, and marginal effects (slopes). Find a lot of examples and vignettes online at: easystats.github.io/modelbased/
