How to use the pipe operator in dplyr

[@candicekoolhaas4343]

I hope whatever university you work at is paying you more than enough. Your stats videos are the best and your personality/delivery is awesome.. you make stats fun!

[@rayray0313]

This is great stuff. Thanks for sharing!

[@Fringer15]

Can you do a Video about Multilevel Regression in R? package(lme4)? Especially about the way to test robustness and things like R^2? All the best 🤙🏼

[@pipertripp]

R now has a native pipe (R 4.1+), |> which you can setup in R Studio under Global Options -> Code -> Editing (it's a checkbox, 4th from top). I suppose the only real advantage of it is that you don't need to import the magrittr package to use pipes now. I always use keyboard short cut (the one you mentioned) so no savings on typing, but the native pipe is represented by fewer characters, for whatever it's worth.

[@QuantPsych]

I've tried switching to the native pipe, but have two problems with it: (1) it requires R > 4.1, which would really screw up some users of flexplot, and (2) if I remember right, there's no way to reference the dataset as there is with magrittr's pipe (e.g., a %>% function(p, q, .)).

[@pipertripp]

@@QuantPsych ah, interesting about '.', I just tried it myself and confirmed what you're saying. gss_cat |> lm(rincome ~ age, data=.) fails but gss_cat %>% lm(rincome ~ age, data=.) works just fine (the model won't be winning any nobels or kaggle competitions any time soon, sadly).

[@pipertripp]

I learned that with native pipes, instead of using '.' you use '_' to reference the piped object.