I run the occasional live stream via the streaming platform Twitch where I talk about R, statistics, and whatever else comes up (yes, I am that much of a nerd). Below, you can find the schedule for the upcoming streams, how to join/participate, what topics have been covered in the past (including all the materials), some additional notes and frequently asked questions (FAQs) about these streams, and a list of some future topics I might cover.

Dates are given as year-month-day.¹⁾ Times are given in terms of my local timezone (which is CET/CEST).

Note: The starting times are exact, but the ending times might fluctuate a bit. More details on the topic(s) to be covered in each stream will be added over time. Further dates to be announced.

Calendar Subscription: You can subscribe to the stream calendar using this .ics file (note: you should subscribe to this calendar and not just import it as a static .ics file/calendar, as otherwise you won't get any updates/changes). For people using Google Calendar, this link should also work.

The streams are free and there is no registration. Once the stream goes live (about 5-10 minutes before the actual start time), simply go to this link to start watching.

You do not need a Twitch account to watch the streams, but if you would like to post comments or questions via the chat, then you do need to be signed in. To create an account, go to Twitch, click on "Sign Up", and follow the instructions. Make sure you also verify your email address as part of the registration procedure (otherwise you cannot use the chat).

The materials from previous streams can be downloaded from the links provided above. Alternatively, you can obtain the materials from the streams via this GitHub repo.

Some notes and frequently asked questions that may come up:

• If you want to follow along with what I cover during the streams, you need to install the current version of R, which you can obtain from the Comprehensive R Archive Network (CRAN). Although not strictly necessary, it will also be useful to install an integrated development environment (IDE) for R. A popular choice these days is RStudio. So, unless you already have a different setup, download RStudio from here and install it in the usual manner.

• During the stream, I share the code I am writing via this GitHub repo. The repo is updated every 60 seconds, so this way, you can easily copy-paste code from the script that I am working on and run the code locally. Suggestions for improvements to the code I am writing are always welcome!

• The streams are not recorded. The whole point is that these are interactive live streams, which sets this form of teaching apart from online courses that use pre-recorded videos or any of the thousands of YouTube videos that teach R and statistics. If I wanted to create the latter, I would script the entire video (which is not compatible with the dynamics of an interactive live stream) and do a lot of post-processing. If you are looking for such online courses, you could look into those that are offered at Coursera, edX, and similar platforms. Note though that the materials from the streams are available, so if you are interested in what was covered, you can always take a look at those.

• I use Twitch for the live streams because it is free to use, can handle an essentially unlimited number of viewers, and does not require the installation of additional software (i.e., streams can be watched through your browser). However, Twitch does need to keep their servers running to provide this service and they do so in part by showing viewers the occasional advertisement. I am not crazy about this, but it's a relatively small inconvenience given the advantages of the platform. And if you do get an ad, then this should only happen at the beginning when you join the stream (unless you take active measures to block ads via software that does so).

Below is a list of potential future topics to be covered during the streams. If you have any suggestions for any further topics, feel free to get in touch or just let me know during one of the streams.

• IDEs for R (i.e., alternatives to RStudio, such as RKWard and Architect, but also more general programming editors such as Vim, Emacs / ESS, Visual Studio Code / VSCodium, Sublime Text, Notepad++ / Tinn-R)
• GUIs for R (jamovi, JASP, Rcmdr, BlueSky, see also r4stats; already did a bit of this on 2020-11-19)
• some potentially interesting books to cover (requirement: book must be freely available online):
  • Gelman et al. (2014). Bayesian data analysis. [link]
  • Chollet & Allaire (2018). Deep learning with R. [link]
  • Goodfellow et al. (2016). Deep learning. [link]
  • Prince (2023). Understanding deep learning. [link]
  • van Buuren (2012). Flexible imputation of missing data. [link]
  • Baumer et al. (2021). Modern data science with R. [link]
  • Çetinkaya-Rundel & Johanna Hardin (2021). Introduction to modern statistics. [link], [link]
  • Bryan (?). STAT 545: Data wrangling, exploration, and analysis with R. [link]
  • Rodrigues (2023). Building reproducible analytical pipelines with R. [link]
  • Albano (2020). Introduction to educational and psychological measurement using R. [link]
  • Grolemund (2014). Hands-on programming with R. [link]
  • Wickham et al. (2023). R for data science. [link]
  • Bryan (?). Happy Git and GitHub for the useR. [link]
  • Baumer et al. (2023). Modern data science with R (3rd ed.). [link]
  • Healy, K. (2019). The plain person's guide to plain text social science. [link]
  • Gagolewski, M. (2023). Deep R programming. [link]
  • Douglas et al. (2023). An introduction to R. [link]
  • Thulin (2021). Modern Statistics with R. [link]
• tutorials / other materials:
  • Bryan (?). What they forgot to teach you about R. [link]
  • Harrell (2023). R workflow for reproducible data analysis and reporting. [link]
  • Matloff (2023). fasteR: Fast lane to learning R! [link]
  • Matloff (2023). Teaching R in a kinder, gentler, more effective manner: Teach base-R, not just the tidyverse. [link]
  • Koncevičius (2022). R base plotting without wrappers. [link]
  • McCarthy (2021/2022). PSYC 615 and PSYC 617.
• syntax style guides and syntax formatting with packages like styler and lintr
• give meta-analysis in SPSS 28+ a try and compare to R/metafor
• variable and value labels in R (see here, tinylabels, vtable, etc.)
• creating tables (see here and the tinytable package which looks very interesting)
• ESM/EMA data analysis (esp. the two chapters I wrote for this book)
• more complex regression models with multiple predictors and non-linearities; combine this with data splitting and cross-validation
• fitting distributions to data / testing the fit between distributions and data
• mixture models / clustering
• complex ANOVAs (ez package)
• nonlinear regression (see data from 2021-03-04)
• Type I/II/III tests
• CFA/SEM with lavaan (already did some CFA on 2020-10-15)
• item response theory
• time series analysis
• multimodel inference / model averaging
• network analysis
• optimization / root finding / integration (did a bunch of optimization stuff already when covering the ISL book)
• GAMs (see this course by Noam Ross)
• maximum likelihood estimation (start with binomial distribution, then normal, regression models; review optimization and computing the Hessian analytically/numerically; nice visualization)
• sparse matrices (Matrix package)
• common statistical tests are linear models
• dealing with date objects / variables
• data.table package
• review *apply() functions
• dplyr (alternatively: poorman)
• multicore / parallel processing
• using other math / linear algebra routines (e.g., OpenBLAS, MKL, ATLAS, BLIS) than the ones that ship with R
• cloud computing
• try out some current code on S/S-PLUS
• version control and especially Git + GitHub/GitLab
• project / file organization (e.g., how to name files) / analysis workflows
• packages for reproducibility (renv, checkpoint, groundhog, …)
• Docker / Rocker; also relevant: Wiebels & Moreau, 2021, Peikert & Brandmaier, 2021, rang
• more on plotting (base-R graphics, ggplot2, lattice, …)
• S3 (generics, methods, etc.), S4, S7
• R Markdown
• Shiny
• robust regression (e.g., MASS::lqs(), MASS::rlm())