geom boxplot doesn't require x aes!

[dannyparsons]

A very minor changed in ggplot2 v3.0.0 is that this now works: ggplot(mtcars, aes(y = disp)) + geom_boxplot() So x isn't required. This could simplify some of our code, although I don't like that the graph shows a scale for the x axis which is confusing as it doesn't really have any meaning: Compared to ggplot(mtcars, aes(x = "", y = disp)) + geom_boxplot() which is simpler.

I'm not sure we need to change anything but it's good to know x is now optional. This is much better for teaching ggplot2 as it wasn't obvious to explain why x is needed for boxplots.

[maxwellfundi]

i agree this looks simpler. Possibly there could be a way to get rid of the scales when we have no x

[dannyparsons]

I've asked about this on the original ggplot2 issue so will see what the response is to that.

[maxwellfundi]

Great! Regards Maxwell Fundi || Phone :+254728508241||

On Mon, Jul 23, 2018 at 3:20 PM Danny Parsons notifications@github.com wrote:

I've asked about this on the original ggplot2 issue so will see what the response is to that.

— You are receiving this because you were assigned. Reply to this email directly, view it on GitHub https://github.com/africanmathsinitiative/R-Instat/issues/4807#issuecomment-407038765, or mute the thread https://github.com/notifications/unsubscribe-auth/AJWEIQIHvsFQDC1W2ECnV4rgKdKZ0Mqeks5uJb90gaJpZM4Va3f8 .

[Ogik99]

@maxwellfundi, @dannyparsons any updates on this

[dannyparsons]

The issue on ggplot2's repo was closed so I'm not sure if anything was done. Someone could check with the latest version of ggplot2 what happens with ggplot(mtcars, aes(y = disp)) + geom_boxplot() now.

[Ogik99]

The scale on the x axis still exists @dannyparsons

[dannyparsons]

Ok thanks, I think we then leave as we have it until it improves by having the x scales removed.