Chapter 8 Describing data sets
When it comes to the description of data sets, it’s often about just getting a quick feel for the data. While dplyr’s summarize() is a great weapon for summarizing singular variables – and I have already shown you in Chapter @(wrangling) how to do this – you need a hassle-free tool to quickly get an overview. For this use-case, skimr is my personal weapon of choice. It is not part of the tidyverse – yet super-compatible – so we first need to load (and perhaps install) it.
if (!"skimr" %in% installed.packages()[,1]) install.packages("skimr")
library(skimr)
library(tidyverse)8.1 Basic descriptives
skimr is designed around its main function skim(). This function handles tibbles as well as vectors. It returns a so-called skim_df which is basically a tibble with some added columns. However, its biggest strength is the fact that it provides descriptives fast and in an easily comprehensible manner. Moreover, you can then go on and further modify them using tidyverse function.
skim(mtcars) # it excels with tibbles as input## ── Data Summary ────────────────────────
## Values
## Name mtcars
## Number of rows 32
## Number of columns 11
## _______________________
## Column type frequency:
## numeric 11
## ________________________
## Group variables None
##
## ── Variable type: numeric ──────────────────────────────────────────────────────
## skim_variable n_missing complete_rate mean sd p0 p25 p50
## 1 mpg 0 1 20.1 6.03 10.4 15.4 19.2
## 2 cyl 0 1 6.19 1.79 4 4 6
## 3 disp 0 1 231. 124. 71.1 121. 196.
## 4 hp 0 1 147. 68.6 52 96.5 123
## 5 drat 0 1 3.60 0.535 2.76 3.08 3.70
## 6 wt 0 1 3.22 0.978 1.51 2.58 3.32
## 7 qsec 0 1 17.8 1.79 14.5 16.9 17.7
## 8 vs 0 1 0.438 0.504 0 0 0
## 9 am 0 1 0.406 0.499 0 0 0
## 10 gear 0 1 3.69 0.738 3 3 4
## 11 carb 0 1 2.81 1.62 1 2 2
## p75 p100 hist
## 1 22.8 33.9 ▃▇▅▁▂
## 2 8 8 ▆▁▃▁▇
## 3 326 472 ▇▃▃▃▂
## 4 180 335 ▇▇▆▃▁
## 5 3.92 4.93 ▇▃▇▅▁
## 6 3.61 5.42 ▃▃▇▁▂
## 7 18.9 22.9 ▃▇▇▂▁
## 8 1 1 ▇▁▁▁▆
## 9 1 1 ▇▁▁▁▆
## 10 4 5 ▇▁▆▁▂
## 11 4 8 ▇▂▅▁▁skim(mtcars$mpg) # vectors are fine, too## ── Data Summary ────────────────────────
## Values
## Name mtcars$mpg
## Number of rows 32
## Number of columns 1
## _______________________
## Column type frequency:
## numeric 1
## ________________________
## Group variables None
##
## ── Variable type: numeric ──────────────────────────────────────────────────────
## skim_variable n_missing complete_rate mean sd p0 p25 p50 p75
## 1 data 0 1 20.1 6.03 10.4 15.4 19.2 22.8
## p100 hist
## 1 33.9 ▃▇▅▁▂Usually, data sets come with columns in different flavors. As an example, we can look at the IMDb data set. This set contains categorical and numeric variables:
imdb_raw <- read_csv("data/imdb2006-2016.csv")## Rows: 1000 Columns: 12## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (5): Title, Genre, Description, Director, Actors
## dbl (7): Rank, Year, Runtime (Minutes), Rating, Votes, Revenue (Millions), M...##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.imdb_raw %>% skim()## ── Data Summary ────────────────────────
## Values
## Name Piped data
## Number of rows 1000
## Number of columns 12
## _______________________
## Column type frequency:
## character 5
## numeric 7
## ________________________
## Group variables None
##
## ── Variable type: character ────────────────────────────────────────────────────
## skim_variable n_missing complete_rate min max empty n_unique whitespace
## 1 Title 0 1 2 61 0 999 0
## 2 Genre 0 1 5 26 0 207 0
## 3 Description 0 1 42 421 0 1000 0
## 4 Director 0 1 3 32 0 644 0
## 5 Actors 0 1 43 77 0 996 0
##
## ── Variable type: numeric ──────────────────────────────────────────────────────
## skim_variable n_missing complete_rate mean sd p0 p25
## 1 Rank 0 1 500. 289. 1 251.
## 2 Year 0 1 2013. 3.21 2006 2010
## 3 Runtime (Minutes) 0 1 113. 18.8 66 100
## 4 Rating 0 1 6.72 0.945 1.9 6.2
## 5 Votes 0 1 169808. 188763. 61 36309
## 6 Revenue (Millions) 128 0.872 83.0 103. 0 13.3
## 7 Metascore 64 0.936 59.0 17.2 11 47
## p50 p75 p100 hist
## 1 500. 750. 1000 ▇▇▇▇▇
## 2 2014 2016 2016 ▃▂▂▃▇
## 3 111 123 191 ▂▇▅▁▁
## 4 6.8 7.4 9 ▁▁▃▇▃
## 5 110799 239910. 1791916 ▇▁▁▁▁
## 6 48.0 114. 937. ▇▁▁▁▁
## 7 59.5 72 100 ▁▅▇▇▂Of course, the type of variable determines the type of operation that can be performed and skimr smartly distinguishes between variable types.
8.1.1 Grouped descriptives
You can look at certain descriptives by group. Then, skim() will return descriptives for each group in a row-wise fashion. Just call group_by() before passing the tibble to the skim() call
mtcars %>%
group_by(cyl) %>%
skim()## ── Data Summary ────────────────────────
## Values
## Name Piped data
## Number of rows 32
## Number of columns 11
## _______________________
## Column type frequency:
## numeric 10
## ________________________
## Group variables cyl
##
## ── Variable type: numeric ──────────────────────────────────────────────────────
## skim_variable cyl n_missing complete_rate mean sd p0 p25
## 1 mpg 4 0 1 26.7 4.51 21.4 22.8
## 2 mpg 6 0 1 19.7 1.45 17.8 18.6
## 3 mpg 8 0 1 15.1 2.56 10.4 14.4
## 4 disp 4 0 1 105. 26.9 71.1 78.8
## 5 disp 6 0 1 183. 41.6 145 160
## 6 disp 8 0 1 353. 67.8 276. 302.
## 7 hp 4 0 1 82.6 20.9 52 65.5
## 8 hp 6 0 1 122. 24.3 105 110
## 9 hp 8 0 1 209. 51.0 150 176.
## 10 drat 4 0 1 4.07 0.365 3.69 3.81
## 11 drat 6 0 1 3.59 0.476 2.76 3.35
## 12 drat 8 0 1 3.23 0.372 2.76 3.07
## 13 wt 4 0 1 2.29 0.570 1.51 1.88
## 14 wt 6 0 1 3.12 0.356 2.62 2.82
## 15 wt 8 0 1 4.00 0.759 3.17 3.53
## 16 qsec 4 0 1 19.1 1.68 16.7 18.6
## 17 qsec 6 0 1 18.0 1.71 15.5 16.7
## 18 qsec 8 0 1 16.8 1.20 14.5 16.1
## 19 vs 4 0 1 0.909 0.302 0 1
## 20 vs 6 0 1 0.571 0.535 0 0
## 21 vs 8 0 1 0 0 0 0
## 22 am 4 0 1 0.727 0.467 0 0.5
## 23 am 6 0 1 0.429 0.535 0 0
## 24 am 8 0 1 0.143 0.363 0 0
## 25 gear 4 0 1 4.09 0.539 3 4
## 26 gear 6 0 1 3.86 0.690 3 3.5
## 27 gear 8 0 1 3.29 0.726 3 3
## 28 carb 4 0 1 1.55 0.522 1 1
## 29 carb 6 0 1 3.43 1.81 1 2.5
## 30 carb 8 0 1 3.5 1.56 2 2.25
## p50 p75 p100 hist
## 1 26 30.4 33.9 ▇▃▂▃▃
## 2 19.7 21 21.4 ▅▂▂▁▇
## 3 15.2 16.2 19.2 ▂▁▇▃▂
## 4 108 121. 147. ▇▂▂▆▃
## 5 168. 196. 258 ▇▁▁▂▂
## 6 350. 390 472 ▇▅▃▂▅
## 7 91 96 113 ▃▆▁▇▃
## 8 110 123 175 ▇▃▁▁▂
## 9 192. 241. 335 ▇▂▃▁▁
## 10 4.08 4.16 4.93 ▇▅▃▁▂
## 11 3.9 3.91 3.92 ▂▂▁▂▇
## 12 3.12 3.22 4.22 ▃▇▁▁▁
## 13 2.2 2.62 3.19 ▇▅▇▂▅
## 14 3.22 3.44 3.46 ▅▂▁▂▇
## 15 3.76 4.01 5.42 ▇▇▁▁▃
## 16 18.9 20.0 22.9 ▃▇▇▁▂
## 17 18.3 19.2 20.2 ▃▇▃▃▇
## 18 17.2 17.6 18 ▂▂▁▅▇
## 19 1 1 1 ▁▁▁▁▇
## 20 1 1 1 ▆▁▁▁▇
## 21 0 0 0 ▁▁▇▁▁
## 22 1 1 1 ▃▁▁▁▇
## 23 0 1 1 ▇▁▁▁▆
## 24 0 0 1 ▇▁▁▁▁
## 25 4 4 5 ▁▁▇▁▂
## 26 4 4 5 ▃▁▇▁▂
## 27 3 3 5 ▇▁▁▁▁
## 28 2 2 2 ▇▁▁▁▇
## 29 4 4 6 ▃▁▇▁▂
## 30 3.5 4 8 ▇▇▁▁▁8.1.2 Modifying the skim() output
You can also limit the variables that should be described in the call:
mtcars %>%
group_by(cyl) %>%
skim(mpg, hp, wt)## ── Data Summary ────────────────────────
## Values
## Name Piped data
## Number of rows 32
## Number of columns 11
## _______________________
## Column type frequency:
## numeric 3
## ________________________
## Group variables cyl
##
## ── Variable type: numeric ──────────────────────────────────────────────────────
## skim_variable cyl n_missing complete_rate mean sd p0 p25 p50
## 1 mpg 4 0 1 26.7 4.51 21.4 22.8 26
## 2 mpg 6 0 1 19.7 1.45 17.8 18.6 19.7
## 3 mpg 8 0 1 15.1 2.56 10.4 14.4 15.2
## 4 hp 4 0 1 82.6 20.9 52 65.5 91
## 5 hp 6 0 1 122. 24.3 105 110 110
## 6 hp 8 0 1 209. 51.0 150 176. 192.
## 7 wt 4 0 1 2.29 0.570 1.51 1.88 2.2
## 8 wt 6 0 1 3.12 0.356 2.62 2.82 3.22
## 9 wt 8 0 1 4.00 0.759 3.17 3.53 3.76
## p75 p100 hist
## 1 30.4 33.9 ▇▃▂▃▃
## 2 21 21.4 ▅▂▂▁▇
## 3 16.2 19.2 ▂▁▇▃▂
## 4 96 113 ▃▆▁▇▃
## 5 123 175 ▇▃▁▁▂
## 6 241. 335 ▇▂▃▁▁
## 7 2.62 3.19 ▇▅▇▂▅
## 8 3.44 3.46 ▅▂▁▂▇
## 9 4.01 5.42 ▇▇▁▁▃When you transform the skim_df into a normal tibble, you can see that additional columns are added:
mtcars %>%
group_by(cyl) %>%
skim(mpg, hp, wt) %>%
as_tibble()## # A tibble: 9 × 13
## skim_type skim_variable cyl n_missing complete_rate numeric.mean numeric.sd
## <chr> <chr> <dbl> <int> <dbl> <dbl> <dbl>
## 1 numeric mpg 4 0 1 26.7 4.51
## 2 numeric mpg 6 0 1 19.7 1.45
## 3 numeric mpg 8 0 1 15.1 2.56
## 4 numeric hp 4 0 1 82.6 20.9
## 5 numeric hp 6 0 1 122. 24.3
## 6 numeric hp 8 0 1 209. 51.0
## 7 numeric wt 4 0 1 2.29 0.570
## 8 numeric wt 6 0 1 3.12 0.356
## 9 numeric wt 8 0 1 4.00 0.759
## # … with 6 more variables: numeric.p0 <dbl>, numeric.p25 <dbl>,
## # numeric.p50 <dbl>, numeric.p75 <dbl>, numeric.p100 <dbl>,
## # numeric.hist <chr>Those are: skim_type which denotes the type of variable (usually “categorical” or “numeric”) and skim_variable containing the name of the variable that is summarized. If there are grouping variables, those are included with their original name (in this case, the number of cylinder, “cyl”).
Therefore, normal dplyr syntax works in a pipeline with a skim_df:
mtcars %>%
group_by(cyl) %>%
skim(mpg, hp, wt) %>%
select(skim_type:numeric.sd)## ── Data Summary ────────────────────────
## Values
## Name Piped data
## Number of rows 32
## Number of columns 11
## _______________________
## Column type frequency:
## numeric 3
## ________________________
## Group variables cyl
##
## ── Variable type: numeric ──────────────────────────────────────────────────────
## skim_variable cyl n_missing complete_rate mean sd
## 1 mpg 4 0 1 26.7 4.51
## 2 mpg 6 0 1 19.7 1.45
## 3 mpg 8 0 1 15.1 2.56
## 4 hp 4 0 1 82.6 20.9
## 5 hp 6 0 1 122. 24.3
## 6 hp 8 0 1 209. 51.0
## 7 wt 4 0 1 2.29 0.570
## 8 wt 6 0 1 3.12 0.356
## 9 wt 8 0 1 4.00 0.759There are also two handy shortcuts to distinguish between the different types of data. partition() takes a skim_df object and returns a list containing tibbles for each variable type. yank() can be used to deliberately choose a variable type upfront.
imdb_raw %>%
skim() %>%
partition() ## $character
##
## ── Variable type: character ────────────────────────────────────────────────────
## skim_variable n_missing complete_rate min max empty n_unique whitespace
## 1 Title 0 1 2 61 0 999 0
## 2 Genre 0 1 5 26 0 207 0
## 3 Description 0 1 42 421 0 1000 0
## 4 Director 0 1 3 32 0 644 0
## 5 Actors 0 1 43 77 0 996 0
##
## $numeric
##
## ── Variable type: numeric ──────────────────────────────────────────────────────
## skim_variable n_missing complete_rate mean sd p0 p25
## 1 Rank 0 1 500. 289. 1 251.
## 2 Year 0 1 2013. 3.21 2006 2010
## 3 Runtime (Minutes) 0 1 113. 18.8 66 100
## 4 Rating 0 1 6.72 0.945 1.9 6.2
## 5 Votes 0 1 169808. 188763. 61 36309
## 6 Revenue (Millions) 128 0.872 83.0 103. 0 13.3
## 7 Metascore 64 0.936 59.0 17.2 11 47
## p50 p75 p100 hist
## 1 500. 750. 1000 ▇▇▇▇▇
## 2 2014 2016 2016 ▃▂▂▃▇
## 3 111 123 191 ▂▇▅▁▁
## 4 6.8 7.4 9 ▁▁▃▇▃
## 5 110799 239910. 1791916 ▇▁▁▁▁
## 6 48.0 114. 937. ▇▁▁▁▁
## 7 59.5 72 100 ▁▅▇▇▂imdb_raw %>%
skim() %>%
yank("numeric")##
## ── Variable type: numeric ──────────────────────────────────────────────────────
## skim_variable n_missing complete_rate mean sd p0 p25
## 1 Rank 0 1 500. 289. 1 251.
## 2 Year 0 1 2013. 3.21 2006 2010
## 3 Runtime (Minutes) 0 1 113. 18.8 66 100
## 4 Rating 0 1 6.72 0.945 1.9 6.2
## 5 Votes 0 1 169808. 188763. 61 36309
## 6 Revenue (Millions) 128 0.872 83.0 103. 0 13.3
## 7 Metascore 64 0.936 59.0 17.2 11 47
## p50 p75 p100 hist
## 1 500. 750. 1000 ▇▇▇▇▇
## 2 2014 2016 2016 ▃▂▂▃▇
## 3 111 123 191 ▂▇▅▁▁
## 4 6.8 7.4 9 ▁▁▃▇▃
## 5 110799 239910. 1791916 ▇▁▁▁▁
## 6 48.0 114. 937. ▇▁▁▁▁
## 7 59.5 72 100 ▁▅▇▇▂8.1.3 Further descriptives
Note that skimr only supports a limited number of measures. If you want to add further descriptives, you can use dplyr::summarize() combined with the across() function to compute the measures. Then you would have to spread the tibbles using tidyr::pivot_longer() and tidyr::pivot_wider() and, finally, join together the tibbles. For the imdb data set and the measures median and variance, this would look like as follows:
imdb_raw %>%
summarize(across(where(is.numeric),
list(
median = ~median(.x, na.rm = TRUE),
variance = ~var(.x, na.rm = TRUE)
))) %>%
pivot_longer(
cols = everything(),
names_to = c("skim_variable", ".value"),
names_sep = "_"
) %>%
right_join(imdb_raw %>% skim() %>% yank("numeric"))## Joining, by = "skim_variable"## # A tibble: 7 × 13
## skim_variable median variance n_missing complete_rate mean sd p0
## <chr> <dbl> <dbl> <int> <dbl> <dbl> <dbl> <dbl>
## 1 Rank 5.00e2 8.34e+ 4 0 1 5.00e2 2.89e+2 1
## 2 Year 2.01e3 1.03e+ 1 0 1 2.01e3 3.21e+0 2006
## 3 Runtime (Minut… 1.11e2 3.54e+ 2 0 1 1.13e2 1.88e+1 66
## 4 Rating 6.8 e0 8.94e- 1 0 1 6.72e0 9.45e-1 1.9
## 5 Votes 1.11e5 3.56e+10 0 1 1.70e5 1.89e+5 61
## 6 Revenue (Milli… 4.80e1 1.07e+ 4 128 0.872 8.30e1 1.03e+2 0
## 7 Metascore 5.95e1 2.96e+ 2 64 0.936 5.90e1 1.72e+1 11
## # … with 5 more variables: p25 <dbl>, p50 <dbl>, p75 <dbl>, p100 <dbl>,
## # hist <chr>8.2 Communicating results
By default, when used in RMarkdown documents, skimr outputs quite decent looking tables which look like this:
imdb_raw %>%
skim() %>%
yank("numeric")Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 | hist |
|---|---|---|---|---|---|---|---|---|---|---|
| Rank | 0 | 1.00 | 500.50 | 288.82 | 1.0 | 250.75 | 500.50 | 750.25 | 1000.00 | ▇▇▇▇▇ |
| Year | 0 | 1.00 | 2012.78 | 3.21 | 2006.0 | 2010.00 | 2014.00 | 2016.00 | 2016.00 | ▃▂▂▃▇ |
| Runtime (Minutes) | 0 | 1.00 | 113.17 | 18.81 | 66.0 | 100.00 | 111.00 | 123.00 | 191.00 | ▂▇▅▁▁ |
| Rating | 0 | 1.00 | 6.72 | 0.95 | 1.9 | 6.20 | 6.80 | 7.40 | 9.00 | ▁▁▃▇▃ |
| Votes | 0 | 1.00 | 169808.26 | 188762.65 | 61.0 | 36309.00 | 110799.00 | 239909.75 | 1791916.00 | ▇▁▁▁▁ |
| Revenue (Millions) | 128 | 0.87 | 82.96 | 103.25 | 0.0 | 13.27 | 47.98 | 113.72 | 936.63 | ▇▁▁▁▁ |
| Metascore | 64 | 0.94 | 58.99 | 17.19 | 11.0 | 47.00 | 59.50 | 72.00 | 100.00 | ▁▅▇▇▂ |
As you may have noticed, the output looked differently before. This is because, in the RMarkdown chunks above, I set my chunk options to render = knitr::normal_print to avoid the printing of the decent looking table but rather the raw output.
8.2.1 Tables with knitr::kable() and kableExtra
The problem with the default output is the lack of modifiability. The column names, for instance, would suffice for small reports but are inappropriate for something you would want to hand in somewhere or even publish. In RMarkdown, the proper tool for modifying and printing tables is the knitr::kable() function with extended functionalities from the kableExtra package. In the following, I will provide a brief and coarse introduction to the package. It is so coarse that you basically will have to work through the vignettes yourself as for the tables everything is relevant and I am not willing to copy the entire vignettes. So, hit the internet and look at the kableExtra vignette for html and pdf.
I will exemplify how it can look like to make a table with descriptives in an HTML-document. I will also add a description for how you can get this into a Word or \(\LaTeX\) document so that you are not limited to RMarkdown when reporting results.
The kableExtra package needs to be loaded (or even installed) first:
if (!"kableExtra" %in% installed.packages()[,1]) install.packages("kableExtra")
library(kableExtra)##
## Attaching package: 'kableExtra'## The following object is masked from 'package:dplyr':
##
## group_rowsThe basic command is kable(). It takes a tibble and outputs it as an html-table in the so-called Twitter bootstrap style.
mtcars %>% kable()| mpg | cyl | disp | hp | drat | wt | qsec | vs | am | gear | carb | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Mazda RX4 | 21.0 | 6 | 160.0 | 110 | 3.90 | 2.620 | 16.46 | 0 | 1 | 4 | 4 |
| Mazda RX4 Wag | 21.0 | 6 | 160.0 | 110 | 3.90 | 2.875 | 17.02 | 0 | 1 | 4 | 4 |
| Datsun 710 | 22.8 | 4 | 108.0 | 93 | 3.85 | 2.320 | 18.61 | 1 | 1 | 4 | 1 |
| Hornet 4 Drive | 21.4 | 6 | 258.0 | 110 | 3.08 | 3.215 | 19.44 | 1 | 0 | 3 | 1 |
| Hornet Sportabout | 18.7 | 8 | 360.0 | 175 | 3.15 | 3.440 | 17.02 | 0 | 0 | 3 | 2 |
| Valiant | 18.1 | 6 | 225.0 | 105 | 2.76 | 3.460 | 20.22 | 1 | 0 | 3 | 1 |
| Duster 360 | 14.3 | 8 | 360.0 | 245 | 3.21 | 3.570 | 15.84 | 0 | 0 | 3 | 4 |
| Merc 240D | 24.4 | 4 | 146.7 | 62 | 3.69 | 3.190 | 20.00 | 1 | 0 | 4 | 2 |
| Merc 230 | 22.8 | 4 | 140.8 | 95 | 3.92 | 3.150 | 22.90 | 1 | 0 | 4 | 2 |
| Merc 280 | 19.2 | 6 | 167.6 | 123 | 3.92 | 3.440 | 18.30 | 1 | 0 | 4 | 4 |
| Merc 280C | 17.8 | 6 | 167.6 | 123 | 3.92 | 3.440 | 18.90 | 1 | 0 | 4 | 4 |
| Merc 450SE | 16.4 | 8 | 275.8 | 180 | 3.07 | 4.070 | 17.40 | 0 | 0 | 3 | 3 |
| Merc 450SL | 17.3 | 8 | 275.8 | 180 | 3.07 | 3.730 | 17.60 | 0 | 0 | 3 | 3 |
| Merc 450SLC | 15.2 | 8 | 275.8 | 180 | 3.07 | 3.780 | 18.00 | 0 | 0 | 3 | 3 |
| Cadillac Fleetwood | 10.4 | 8 | 472.0 | 205 | 2.93 | 5.250 | 17.98 | 0 | 0 | 3 | 4 |
| Lincoln Continental | 10.4 | 8 | 460.0 | 215 | 3.00 | 5.424 | 17.82 | 0 | 0 | 3 | 4 |
| Chrysler Imperial | 14.7 | 8 | 440.0 | 230 | 3.23 | 5.345 | 17.42 | 0 | 0 | 3 | 4 |
| Fiat 128 | 32.4 | 4 | 78.7 | 66 | 4.08 | 2.200 | 19.47 | 1 | 1 | 4 | 1 |
| Honda Civic | 30.4 | 4 | 75.7 | 52 | 4.93 | 1.615 | 18.52 | 1 | 1 | 4 | 2 |
| Toyota Corolla | 33.9 | 4 | 71.1 | 65 | 4.22 | 1.835 | 19.90 | 1 | 1 | 4 | 1 |
| Toyota Corona | 21.5 | 4 | 120.1 | 97 | 3.70 | 2.465 | 20.01 | 1 | 0 | 3 | 1 |
| Dodge Challenger | 15.5 | 8 | 318.0 | 150 | 2.76 | 3.520 | 16.87 | 0 | 0 | 3 | 2 |
| AMC Javelin | 15.2 | 8 | 304.0 | 150 | 3.15 | 3.435 | 17.30 | 0 | 0 | 3 | 2 |
| Camaro Z28 | 13.3 | 8 | 350.0 | 245 | 3.73 | 3.840 | 15.41 | 0 | 0 | 3 | 4 |
| Pontiac Firebird | 19.2 | 8 | 400.0 | 175 | 3.08 | 3.845 | 17.05 | 0 | 0 | 3 | 2 |
| Fiat X1-9 | 27.3 | 4 | 79.0 | 66 | 4.08 | 1.935 | 18.90 | 1 | 1 | 4 | 1 |
| Porsche 914-2 | 26.0 | 4 | 120.3 | 91 | 4.43 | 2.140 | 16.70 | 0 | 1 | 5 | 2 |
| Lotus Europa | 30.4 | 4 | 95.1 | 113 | 3.77 | 1.513 | 16.90 | 1 | 1 | 5 | 2 |
| Ford Pantera L | 15.8 | 8 | 351.0 | 264 | 4.22 | 3.170 | 14.50 | 0 | 1 | 5 | 4 |
| Ferrari Dino | 19.7 | 6 | 145.0 | 175 | 3.62 | 2.770 | 15.50 | 0 | 1 | 5 | 6 |
| Maserati Bora | 15.0 | 8 | 301.0 | 335 | 3.54 | 3.570 | 14.60 | 0 | 1 | 5 | 8 |
| Volvo 142E | 21.4 | 4 | 121.0 | 109 | 4.11 | 2.780 | 18.60 | 1 | 1 | 4 | 2 |
8.2.2 Use case: reporting skimr results
8.2.2.1 Basics for HTML and PDF
First, I need to get the results in a proper tibble. I use the mtcars data set.
desc_mtcars <- skim(mtcars) %>% as_tibble()First, I select and rename the variables I deem interesting. To save myself some time, I choose variables based on their position:
for_table <- desc_mtcars %>%
select(Variable = 2,
Mean = 5,
SD = 6,
Minimum = 7,
Maximum = 11) %>%
mutate(across(where(is.numeric), ~round(., 1))) # round the outputNow, I have put it into a tibble that, in theory, can be just printed down as a table (you might wish to change the names of the described variables, but let’s save this for another time; if you’re willing to: case_when() may be your best shot). I proceed with the kable() call. By including kable_styling(), you can make some general specifications. column_spec() allows you to customize individual columns. footnote() gives you the opportunity for footnote customization. Those “kables” are extremely customizable with other functions you will find in the manual I linked below.
for_table %>%
kable(caption = "Example for a table with some descriptives") %>%
kable_styling(
bootstrap_options = "striped", # several design options
full_width = FALSE, # defaults to TRUE
position = "center", # where is it positioned?
fixed_thead = TRUE # whether header is fixed when scrolling through -- only for longer tables
) %>%
column_spec(1, bold = TRUE, border_right = TRUE) %>% # column specifications can be easily modified
footnote(general = "You can add some footnotes with certain signs, too.", # this is how you add a footnote
number = c("Footnote 1; ", "Footnote 2; "),
alphabet = c("Footnote A; ", "Footnote B; "),
symbol = c("Footnote Symbol 1; ", "Footnote Symbol 2"),
general_title = "General: ", number_title = "Type I: ",
alphabet_title = "Type II: ", symbol_title = "Type III: ",
footnote_as_chunk = T, title_format = c("italic", "underline")
)| Variable | Mean | SD | Minimum | Maximum |
|---|---|---|---|---|
| mpg | 20.1 | 6.0 | 10.4 | 33.9 |
| cyl | 6.2 | 1.8 | 4.0 | 8.0 |
| disp | 230.7 | 123.9 | 71.1 | 472.0 |
| hp | 146.7 | 68.6 | 52.0 | 335.0 |
| drat | 3.6 | 0.5 | 2.8 | 4.9 |
| wt | 3.2 | 1.0 | 1.5 | 5.4 |
| qsec | 17.8 | 1.8 | 14.5 | 22.9 |
| vs | 0.4 | 0.5 | 0.0 | 1.0 |
| am | 0.4 | 0.5 | 0.0 | 1.0 |
| gear | 3.7 | 0.7 | 3.0 | 5.0 |
| carb | 2.8 | 1.6 | 1.0 | 8.0 |
| General: You can add some footnotes with certain signs, too. | ||||
| Type I: 1 Footnote 1; 2 Footnote 2; | ||||
| Type II: a Footnote A; b Footnote B; | ||||
| Type III: * Footnote Symbol 1; † Footnote Symbol 2 |
For \(\LaTeX\) tables this would generally look the same. Depending on whether you knit the RMarkdown file to PDF or HTML, the output will change. You also have some different specification options for \(\LaTeX\) output.
8.2.2.2 Getting it into a Word file
Unfortunately, you cannot yet output the tables produced with kable into a proper Word file. In order to achieve this, however, you can use flextable and then export the tables in different formats. flextable is similarly capable as kable. An introduction can be found here.
A coarse example follows.
First, it needs to be loaded.
if (!"flextable" %in% installed.packages()[,1]) install.packages("flextable")
library(flextable)##
## Attaching package: 'flextable'## The following objects are masked from 'package:kableExtra':
##
## as_image, footnote## The following object is masked from 'package:purrr':
##
## composeThen, I can produce the table and export it to DOCX or PPTX format using the save_as_*() function.
for_table %>%
flextable() %>%
save_as_docx(path = "example_table_docx.docx", sep = "/")
for_table %>%
flextable() %>%
save_as_pptx(path = "example_table_pptx.pptx", sep = "/")Then, further manipulations can be performed in MS Word or PowerPoint (or LibreOffice or whatever).
Also, if you have a working version of Word or PowerPoint on your machine, you can use the following commands to start up an interactive session:
for_table %>%
flextable() %>%
print(preview = "docx")
for_table %>%
flextable() %>%
print(preview = "pptx")8.3 Further links
- An exhaustive introduction to
skimr. - Manual for creation of
kabletables in HTML and \(\LaTeX\). - Intro to
flextable. - More on
kable.