Add tests and method for list in describe_distribution

[etiennebacher]

This PR is related to #97. Sometimes we want to compare the distribution of several variables but not necessarily of the whole dataframe, so this PR adds a method to pass a list to describe_distribution(). Some examples below:

> describe_distribution(list(mtcars$mpg, mtcars$cyl))

Variable   |  Mean |   SD |  IQR |          Range | Skewness | Kurtosis |  n | n_Missing
----------------------------------------------------------------------------------------
mtcars$mpg | 20.09 | 6.03 | 7.53 | [10.40, 33.90] |     0.67 |    -0.02 | 32 |         0
mtcars$cyl |  6.19 | 1.79 | 4.00 |   [4.00, 8.00] |    -0.19 |    -1.76 | 32 |         0

> describe_distribution(list(foo = mtcars$mpg, foo2 = mtcars$cyl))

Variable |  Mean |   SD |  IQR |          Range | Skewness | Kurtosis |  n | n_Missing
--------------------------------------------------------------------------------------
foo      | 20.09 | 6.03 | 7.53 | [10.40, 33.90] |     0.67 |    -0.02 | 32 |         0
foo2     |  6.19 | 1.79 | 4.00 |   [4.00, 8.00] |    -0.19 |    -1.76 | 32 |         0

> describe_distribution(list(foo = mtcars$mpg, mtcars$cyl))

Variable   |  Mean |   SD |  IQR |          Range | Skewness | Kurtosis |  n | n_Missing
----------------------------------------------------------------------------------------
foo        | 20.09 | 6.03 | 7.53 | [10.40, 33.90] |     0.67 |    -0.02 | 32 |         0
mtcars$cyl |  6.19 | 1.79 | 4.00 |   [4.00, 8.00] |    -0.19 |    -1.76 | 32 |         0

> x <- list(mtcars$mpg, mtcars$cyl)
> describe_distribution(x)

Variable |  Mean |   SD |  IQR |          Range | Skewness | Kurtosis |  n | n_Missing
--------------------------------------------------------------------------------------
Var_1    | 20.09 | 6.03 | 7.53 | [10.40, 33.90] |     0.67 |    -0.02 | 32 |         0
Var_2    |  6.19 | 1.79 | 4.00 |   [4.00, 8.00] |    -0.19 |    -1.76 | 32 |         0

Note that if the list is stored in an object (such as x above) and if its elements are unnamed, it is not possible to know the elements that created x, hence the "Var_1", "Var_2" as variable names. I also add tests and some doc for this.

There will probably be some things to change so I didn't add details in the NEWS yet. What do you think?

Close #97.

[codecov-commenter]

Codecov Report

Merging #105 (d76719c) into master (c4c484d) will increase coverage by 2.78%.
The diff coverage is 96.66%.

@@            Coverage Diff             @@
##           master     #105      +/-   ##
==========================================
+ Coverage   73.12%   75.90%   +2.78%     
==========================================
  Files          39       40       +1     
  Lines        2013     2042      +29     
==========================================
+ Hits         1472     1550      +78     
+ Misses        541      492      -49     

Impacted Files	Coverage Δ
R/describe_distribution.R	76.55% <96.66%> (+17.89%)	⬆️
R/weighted_mean_median_sd_mad.R	91.89% <0.00%> (ø)
R/utils_standardize_center.R	77.84% <0.00%> (+9.06%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c4c484d...d76719c. Read the comment docs.

[IndrajeetPatil]

Looks good! Just a few minor comments that need to be attended to.

[IndrajeetPatil]

Have you tried the plotting method in {see} and checked if it works with outputs from list input?

Maybe you can post an example in the PR thread.

[etiennebacher]

I didn't know there was a plotting method for that. I just tried plot(describe_distribution(list(mtcars$mpg, mtcars$cyl))) and it doesn't work. I don't have any experience with ggplot2 programming so not sure how to handle this. I'll take a look at see source code later, maybe it's easier than I think

[IndrajeetPatil]

I think you can leave this in for now and create an issue in {see} repository so that we don't forget about it.

[IndrajeetPatil]

Looks good to me! Thanks for working on this (and also for adding exhaustive unit tests!).

I will wait for @strengejacke to have a look at this before I merge.