Create option to use a dot-plot in visualize()

[kzembower]

Create the ability for visualize() to display a dotplot, in addition to a barchart

As a first-year statistics student, I think it would be useful for visualize() to be able to display a dotplot, in addition to a barchart.

I realize that this would just be a cosmetic change, and would not add to the information already displayed by visualize(). However, in my course, we are using the popular statistics textbook, "Statistics: Unlocking the Power of Data" by Lock, Lock, Lock, Lock and Lock (third edition). This textbook makes heavy use of dotplots in their examples of hypothesis tests and computation of confidence intervals.

Even though I don't think it's needed, here's a MWE:

library(infer)

set.seed(1)

df <- data.frame(d = rnorm(100))
str(df)

(xbar <- df %>%
     specify(response = d) %>%
     calculate(stat = "mean"))

null_dist <- df %>%
    specify(response = d) %>%
    hypothesize(null = "point", mu = 0) %>%
    generate(reps = 1000, type = "bootstrap") %>%
    calculate(stat = "mean")

visualize(null_dist) +
    shade_p_value(obs_stat = xbar, direction = "two-tailed")

As I said, this would not add anything of significance to the infer library, but may make it more approachable for beginning statistics students.

Thanks for considering this feature request.

-Kevin