Econometrics and Free Software by Bruno Rodrigues.
RSS feed for blog post updates.
Follow me on Mastodon, twitter, or check out my Github.
Check out my package that adds logging to R functions, {chronicler}.
Or read my free ebooks, to learn some R and build reproducible analytical pipelines..
You can also watch my youtube channel or find the slides to the talks I've given here.
Buy me a coffee, my kids don't let me sleep.

Why I find tidyeval useful


First thing’s first: maybe you shouldn’t care about tidyeval. Maybe you don’t need it. If you exclusively work interactively, I don’t think that learning about tidyeval is important. I can only speak for me, and explain to you why I personally find tidyeval useful.

I wanted to write this blog post after reading this twitter thread and specifically this question.

Mara Averick then wrote this blogpost linking to 6 other blog posts that give some tidyeval examples. Reading them, plus the Programming with dplyr vignette should help you get started with tidyeval.

But maybe now you know how to use it, but not why and when you should use it… Basically, whenever you want to write a function that looks something like this:

my_function(my_data, one_column_inside_data)

is when you want to use the power of tidyeval.

I work at STATEC, Luxembourg’s national institute of statistics. I work on all kinds of different projects, and when data gets updated (for example because a new round of data collection for some survey finished), I run my own scripts on the fresh data to make the data nice and tidy for analysis. Because surveys get updated, sometimes column names change a little bit, and this can cause some issues.

Very recently, a dataset I work with got updated. Data collection was finished, so I just loaded my hombrewed package written for this project, changed the path from last year’s script to this year’s fresh data path, ran the code, and watched as the folders got populated with new ggplot2 graphs and LaTeX tables with descriptive statistics and regression results. This is then used to generate this year’s report. However, by looking at the graphs, I noticed something weird; some graphs were showing some very strange patterns. It turns out that one column got its name changed, and also one of its values got changed too.

Last year, this column, let’s call it spam, had values 1 for good and 0 for bad. This year the column is called Spam and the values are 1 and 2. When I found out that this was the source of the problem, I just had to change the arguments of my functions from

generate_spam_plot(dataset = data2016, column = spam, value = 1)
generate_spam_plot(dataset = data2016, column = spam, value = 0)


generate_spam_plot(dataset = data2017, column = Spam, value = 1)
generate_spam_plot(dataset = data2017, column = Spam, value = 2)

without needing to change anything else. This is why I use tidyeval; without it, writing a function such as genereta_spam_plot would not be easy. It would be possible, but not easy.

If you want to know more about tidyeval and working programmatically with R, I shamelessly invite you to read a book I’ve been working on: It’s still a WIP, but maybe you’ll find it useful. I plan on finishing it by the end of the year, but there’s already some content to keep you busy!