FAQ

Frequently Asked Questions

NoteHow do I install Rflow?

No installation is required. Rflow is a web-based application that runs in your browser.

NoteIs Rflow free to use?

You can sign up for a free Rflow account and use most features, but the numbers of projects and disk quota are limited.

NoteIs Rflow open source?

Rflow is not currently open source. It uses open source R as a backend, with the plumber2 package to run R functions. It is possible that in the future we may offer it as a self-hosted service and fully release all the source code.

NoteIs the R code downloaded from Rflow usable in RStudio, VS Code or Positron IDEs?

Yes, with two minor caveats:

1. First, you may need to do a one-off installation of additional R packages, for example ggplot2, glmmTMB, car or DHARMa, using the install.packages() function in R.
2. Second, you might need to change the working directory, especially in RStudio, to ensure that the script can find your csv data files in read.csv function calls.

Apart from these two minor issues, the R code downloaded from Rflow is fully usable in any R IDE.

NoteCan I use exported Rflow Workflows in other R IDEs?

No. If you want to export your Rflow workflow for use in an IDE such as VS Code, RStudio or Positron, you need to use the “Download R Code” button in the Rflow workflow editor. This will give you a fully reproducible R script that you can run in any R IDE. The “Export” button on the Rflow projects page is for sharing your workflow with specific other Rflow users if you do not want to make it public.

NoteWhen will Rflow Standard be available to purchase?

Subscriptions for Rflow Standard will be available in late 2026. We are currently in beta testing and will announce the official release date on our website.

NoteWhy are both glm and glmmTMB available in Rflow?

The glm function is installed in the core R package and is most widely documented. glmmTMB is more recent, with more advanced features. There are pros and cons to both.

Feature	glm	glmmTMB
CPU speed	Fast	Slower
Available error models	Limited	Numerous
Random effects	No	Yes
Saved model size	Small	Large

NoteCan I use random effects models in Rflow?

Not yet. Rflow gives access to glmmTMB() which fits both linear and generalised linear mixed models. However, the interface is currently limited to fixed effects models. We plan to add random effects models in a future release.

NoteWhy is DHARMa used to assess residuals in Rflow?

The standard base plot.lm() and especially plot.glm() can be difficult to interpret in models with small sample sizes and or non-normal error distributions. The DHARMa package uses simulation to generate residuals that are more interpretable and easier to assess for model fit.

NoteWhy are so few statistical tests available in Rflow?

You might be puzzled that lots of “standard” statistical tests appear to be missing from Rflow. Many statistics packages have separate functions for numerous tests, e.g. paired t-test, one-way ANOVA, two-way ANOVA, Kruskal-Wallis test, etc. In Rflow, we have unified all of these tests into a single framework based on (generalised) linear models. This has the advantage that you can use the same workflow for many different tests, and it is easier to learn and understand.