R Tips and Tricks

This can be useful when you have lots of columns and want to just look at a certain type. Pipe it into `names()` to just get the names of the columns.

  data(swiss)
  swiss <- as_tibble(swiss)
  swiss %>% select(where(is.double))

A base-R version would be something like:

  data(swiss)
  swiss[, sapply(swiss, is.double)]

Here's a quick and easy way to visually compare histograms of a variable before and after a log transformation. For easy repetition, you can string the last 4 lines together with `;` to create a one-liner, or better, turn them into a simple function. (I'm not sure of the relative strengths and weaknesses of the cowplot and patchwork packages, so I've included examples with both. See this stackoverflow question for other approaches.)

  library(cowplot)
  p <- ggplot(hcvvt, aes(x = my_variable))
  p1 <- p + geom_histogram()
  p2 <- p1 + scale_x_log10()
  plot_grid(p1, p2)

  library(patchwork)
  p <- ggplot(hcvvt, aes(x = my_variable))
  p1 <- p + geom_histogram()
  p2 <- p1 + scale_x_log10()
  p1 + p2