search

ASKurz / Experimental-design-and-the-GLMM

MIT License
54 stars 7 forks source link

Consider making a companion R data package #29

Open mccarthy-m-g opened 1 year ago

mccarthy-m-g commented 1 year ago

Right now your plan is to provide “data sets saved as external files for all analyses in the book”, but I was wondering if you’d consider packaging them into an R package instead (or in addition to)?

This would make it easier to write and share examples for the book, since you’d just need to load the R package to access the data rather than dealing with local file paths. It also provides a nice way to document the data sets.

For a familiar example of what this could look like, I recently packaged the open data from Singer and Willet’s (2003) Applied longitudinal data analysis book into an R data package: https://mccarthy-m-g.github.io/alda/

I don’t know if you have experience with package development or publishing on CRAN, so if you’d like help with that I’d be happy to.

ASKurz commented 1 year ago

Hey @mccarthy-m-g, that seems like a nice idea. The difficulty, which it looks like you already anticipated, is I have no experience with making R packages. So it's a big maybe. Thank you for the offer, though. Once this projects gets further along, I'll keep your offer in mind.

mccarthy-m-g commented 1 year ago

No problem! Happy to contribute if I can.

For what it’s worth, R packages that just ship datasets are really simple and I think you’d pick up the workflow quickly even without prior experience. The extra effort involved is mainly just documenting the datasets in a way similar to how you did in your blog post yesterday.