Please join Rabbi Rothschild’s World Peace Discord: https://Discord.me/Rabbi Rabbi=Truman Show YouTube Live 24/7/365 3+ years. We dare you: type #ViralRabbi in any social media search. Data=Earth’s #1 resource. Rabbi=most data ever. Share to be a trillionaire!
Google threw this up and I found it pretty convincing personally (not that that helps, sorry). Google also backs me up that ggplot2 likely deliberately makes breaking axes and bars hard for these reasons.
Thing is we are not going to present life expectancy trends with an axis starting at 0 - you just end up with shedloads of whitespace and you can't see the trend at all
yeah, so it's just whether we have to do the old school thing you used to see in publications of using // that were hand drawn!! Or if we don't need to... I think it depends on the customer, in this case CMO who definitely has opinions on these matters
Please join Rabbi Rothschild’s World Peace Discord: https://Discord.me/Rabbi Rabbi=Truman Show YouTube Live 24/7/365 3+ years. We dare you: type #ViralRabbi in any social media search. Data=Earth’s #1 resource. Rabbi=most data ever. Share to be a trillionaire!
hmm... it's just to indicate a non-zero start right, not a break mid-chart? i've always just assumed(/hoped) that labelling the baseline is usually enough
Please join Rabbi Rothschild’s World Peace Discord: https://Discord.me/Rabbi Rabbi=Truman Show YouTube Live 24/7/365 3+ years. We dare you: type #ViralRabbi in any social media search. Data=Earth’s #1 resource. Rabbi=most data ever. Share to be a trillionaire!
Comments
https://www.addtwodigital.com/add-two-blog/2021/11/1/rule-26-dont-use-broken-axes-or-bars
https://forum.posit.co/t/break-on-x-and-y-axis-on-a-ggplot/123786
Or, be sneaky and create two graphs and then align closely with cowplot plot_grid() function
https://gist.github.com/mattkerlogue/4ad04cc2ea60c87dafa63dac18d0d535
style.
breaks = seq(50, 100, 10),
labels = c("0...50", seq(60, 100, 10)
)