Occasionally it is useful to generate a table of summary statistics
for rows of a dataset, where such rows represent sampling units and and
columns may be categorical or continuous. The excellent R package table1 does exactly
this, and was the inspiration for tablet
.
table1
however is optimized for html; tablet
tries to provide a format-neutral implementation and relies on kableExtra to
handle the rendering. Support for pdf (latex) is of particular interest,
and is illustrated in the companion vignette. Here we convert that
presentation to html for completeness. If you only need html output, you
may prefer the interface, flexibility, and styling options of
table1
.
To support our examples, we load some other packages and in particular locate the melanoma dataset from boot.
For starters, we’ll just coerce two variables to factor to show that they are categorical, and then pass the whole thing to tablet(). Then we forward to as_kable() for rendering (calls kableExtra::kbl and adds some magic).
All (N = 205) |
|
---|---|
status | |
Mean (SD) | 1.79 (0.551) |
Median (range) | 2 (1, 3) |
sex | |
0 | 126 (61.5%) |
1 | 79 (38.5%) |
age | |
Mean (SD) | 52.5 (16.7) |
Median (range) | 54 (4, 95) |
thickness | |
Mean (SD) | 2.92 (2.96) |
Median (range) | 1.94 (0.1, 17.4) |
ulcer | |
0 | 115 (56.1%) |
1 | 90 (43.9%) |
Now we redefine the dataset, supplying metadata almost verbatim from
?melanoma
. This is fairly easy using package
yamlet
. Note that we reverse the authors’ factor order of
1, 0 for ulcer and move status ‘Alive’ to first position.
x <- melanoma
x %<>% decorate('
time: [ Survival Time Since Operation, day ]
status:
- End of Study Patient Status
-
- Alive: 2
- Melanoma Death: 1
- Unrelated Death: 3
sex: [ Sex, [ Male: 1, Female: 0 ]]
age: [ Age at Time of Operation, year ]
year: [ Year of Operation, year ]
thickness: [ Tumor Thickness, mm ]
ulcer: [ Ulceration, [ Absent: 0, Present: 1 ]]
')
x %<>% select(-time, -year)
x %<>% group_by(status)
x %<>% resolve
group_by(status) causes statistics to be summarized in columns by group.
resolve() disambiguates labels, units, and factor levels (actually creating factors where appropriate, such as for sex and ulcer).
Now we pass x to tablet() and as_kable() for a more informative result.
Alive (N = 134) |
Melanoma Death (N = 57) |
Unrelated Death (N = 14) |
All (N = 205) |
|
---|---|---|---|---|
Sex | ||||
Male | 43 (32.1%) | 29 (50.9%) | 7 (50%) | 79 (38.5%) |
Female | 91 (67.9%) | 28 (49.1%) | 7 (50%) | 126 (61.5%) |
Age at Time of Operation (year) | ||||
Mean (SD) | 50 (15.9) | 55.1 (17.9) | 65.3 (10.9) | 52.5 (16.7) |
Median (range) | 52 (4, 84) | 56 (14, 95) | 65 (49, 86) | 54 (4, 95) |
Tumor Thickness (mm) | ||||
Mean (SD) | 2.24 (2.33) | 4.31 (3.57) | 3.72 (3.63) | 2.92 (2.96) |
Median (range) | 1.36 (0.1, 12.9) | 3.54 (0.32, 17.4) | 2.26 (0.16, 12.6) | 1.94 (0.1, 17.4) |
Ulceration | ||||
Absent | 92 (68.7%) | 16 (28.1%) | 7 (50%) | 115 (56.1%) |
Present | 42 (31.3%) | 41 (71.9%) | 7 (50%) | 90 (43.9%) |
Notice that:
If you don’t particularly care for some aspect of the presentation, you can jump in between tablet() and as_kable() to fix things up. For example, if you don’t want the “All” column you can just say
x %>% tablet %>% select(-All) %>% as_kable
.If you only want the the “All” column, you can just remove the group(s):
x %>% ungroup %>% select(-1) %>% tablet %>% as_kable
.By the way, you can also pass all = NULL
to suppress the
‘All’ column.
In tablet(), most columns are the consequences of a grouping variable. Not surprisingly, grouped columns are just a consequence of nested grouping variables. To illustrate, we follow the table1 vignette by adding a grouping variable that groups the two kinds of death.
x %<>% mutate(class = status) # copy the current group
x %<>% modify(class, label = 'class') # change its label
levels(x$status) <- c('Alive','Melanoma','Unrelated') # tweak current group
levels(x$class) <- c(' ', 'Death', 'Death') # cluster groups
x %<>% group_by(class, status) # nest groups
x %>% tablet %>% as_kable # render
Alive (N = 134) |
Melanoma (N = 57) |
Unrelated (N = 14) |
All (N = 205) |
|
---|---|---|---|---|
Sex | ||||
Male | 43 (32.1%) | 29 (50.9%) | 7 (50%) | 79 (38.5%) |
Female | 91 (67.9%) | 28 (49.1%) | 7 (50%) | 126 (61.5%) |
Age at Time of Operation (year) | ||||
Mean (SD) | 50 (15.9) | 55.1 (17.9) | 65.3 (10.9) | 52.5 (16.7) |
Median (range) | 52 (4, 84) | 56 (14, 95) | 65 (49, 86) | 54 (4, 95) |
Tumor Thickness (mm) | ||||
Mean (SD) | 2.24 (2.33) | 4.31 (3.57) | 3.72 (3.63) | 2.92 (2.96) |
Median (range) | 1.36 (0.1, 12.9) | 3.54 (0.32, 17.4) | 2.26 (0.16, 12.6) | 1.94 (0.1, 17.4) |
Ulceration | ||||
Absent | 92 (68.7%) | 16 (28.1%) | 7 (50%) | 115 (56.1%) |
Present | 42 (31.3%) | 41 (71.9%) | 7 (50%) | 90 (43.9%) |
Categorical observations (in principle) and grouping variables are all factors, and are thus transposable. To illustrate, we drop the column group above and instead nest sex within status …
Male (N = 43) |
Female (N = 91) |
Male (N = 29) |
Female (N = 28) |
Male (N = 7) |
Female (N = 7) |
All (N = 205) |
|
---|---|---|---|---|---|---|---|
Age at Time of Operation (year) | |||||||
Mean (SD) | 52.5 (16.9) | 48.8 (15.4) | 53.9 (19.7) | 56.4 (16.2) | 62.4 (11.2) | 68.1 (10.6) | 52.5 (16.7) |
Median (range) | 55 (12, 84) | 49 (4, 77) | 52 (19, 95) | 58 (14, 89) | 64 (49, 76) | 66 (54, 86) | 54 (4, 95) |
Tumor Thickness (mm) | |||||||
Mean (SD) | 2.73 (2.49) | 2.02 (2.22) | 4.63 (3.47) | 3.99 (3.71) | 4.83 (4.19) | 2.6 (2.84) | 2.92 (2.96) |
Median (range) | 1.62 (0.16, 8.38) | 1.29 (0.1, 12.9) | 4.04 (0.81, 14.7) | 3.14 (0.32, 17.4) | 4.84 (0.65, 12.6) | 1.45 (0.16, 8.54) | 1.94 (0.1, 17.4) |
Ulceration | |||||||
Absent | 24 (55.8%) | 68 (74.7%) | 8 (27.6%) | 8 (28.6%) | 4 (57.1%) | 3 (42.9%) | 115 (56.1%) |
Present | 19 (44.2%) | 23 (25.3%) | 21 (72.4%) | 20 (71.4%) | 3 (42.9%) | 4 (57.1%) | 90 (43.9%) |
… or nest ulceration within status …
Absent (N = 92) |
Present (N = 42) |
Absent (N = 16) |
Present (N = 41) |
Absent (N = 7) |
Present (N = 7) |
All (N = 205) |
|
---|---|---|---|---|---|---|---|
Sex | |||||||
Male | 24 (26.1%) | 19 (45.2%) | 8 (50%) | 21 (51.2%) | 4 (57.1%) | 3 (42.9%) | 79 (38.5%) |
Female | 68 (73.9%) | 23 (54.8%) | 8 (50%) | 20 (48.8%) | 3 (42.9%) | 4 (57.1%) | 126 (61.5%) |
Age at Time of Operation (year) | |||||||
Mean (SD) | 49.3 (15.4) | 51.6 (17.1) | 54.9 (19.9) | 55.1 (17.4) | 58.4 (8.66) | 72.1 (8.53) | 52.5 (16.7) |
Median (range) | 50 (4, 83) | 54.5 (12, 84) | 59 (16, 83) | 56 (14, 95) | 56 (49, 71) | 72 (60, 86) | 54 (4, 95) |
Tumor Thickness (mm) | |||||||
Mean (SD) | 1.63 (1.93) | 3.58 (2.58) | 2.7 (3.35) | 4.94 (3.5) | 2.1 (1.93) | 5.34 (4.33) | 2.92 (2.96) |
Median (range) | 1.13 (0.1, 12.9) | 3.06 (0.32, 12.2) | 1.94 (0.32, 14.7) | 4.04 (0.97, 17.4) | 1.45 (0.65, 6.12) | 4.84 (0.16, 12.6) | 1.94 (0.1, 17.4) |
… or where it makes sense, use multiple levels of nesting.
Male (N = 24) |
Female (N = 68) |
Male (N = 19) |
Female (N = 23) |
Male (N = 8) |
Female (N = 8) |
Male (N = 21) |
Female (N = 20) |
Male (N = 4) |
Female (N = 3) |
Male (N = 3) |
Female (N = 4) |
All (N = 205) |
|
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Age at Time of Operation (year) | |||||||||||||
Mean (SD) | 50.4 (17) | 48.9 (14.9) | 55.3 (16.9) | 48.7 (17) | 55.2 (22.2) | 54.6 (18.8) | 53.3 (19.2) | 57 (15.5) | 54.5 (7.14) | 63.7 (8.74) | 73 (2.65) | 71.5 (11.8) | 52.5 (16.7) |
Median (range) | 54 (15, 83) | 49 (4, 77) | 56 (12, 84) | 48 (19, 75) | 56 (27, 83) | 59 (16, 77) | 52 (19, 95) | 58 (14, 89) | 52.5 (49, 64) | 66 (54, 71) | 72 (71, 76) | 70 (60, 86) | 54 (4, 95) |
Tumor Thickness (mm) | |||||||||||||
Mean (SD) | 1.47 (1.72) | 1.69 (2) | 4.32 (2.42) | 2.97 (2.59) | 3.27 (4.68) | 2.14 (1.18) | 5.14 (2.86) | 4.72 (4.13) | 2.42 (2.5) | 1.67 (1.14) | 8.05 (4.02) | 3.3 (3.71) | 2.92 (2.96) |
Median (range) | 0.97 (0.16, 7.09) | 1.29 (0.1, 12.9) | 3.87 (0.81, 8.38) | 1.94 (0.32, 12.2) | 1.78 (0.81, 14.7) | 2.02 (0.32, 3.56) | 4.83 (1.62, 12.9) | 3.54 (0.97, 17.4) | 1.46 (0.65, 6.12) | 1.45 (0.65, 2.9) | 6.76 (4.84, 12.6) | 2.26 (0.16, 8.54) | 1.94 (0.1, 17.4) |
tablet
tries to give rather exhaustive control over
formatting. Much can be achieved by replacing elements of ‘fun’, ‘fac’,
‘num’, and ‘lab’ (see ?tablet.data.frame
). For finer
control, you can replace these entirely. In this example, we will …
ignore categoricals (other than groups) by replacing ‘fac’ with something of length zero,
drop the ‘N =’ header material by substituting in ‘lab’, and
switch to ‘(min - max)’ instead of ‘(min, max)’.
x %<>% group_by(status)
x %>%
tablet(
fac = NULL,
lab ~ name,
`Median (range)` ~ med + ' (' + min + ' - ' + max + ')'
) %>%
as_kable
Alive | Melanoma | Unrelated | All | |
---|---|---|---|---|
Age at Time of Operation (year) | ||||
Mean (SD) | 50 (15.9) | 55.1 (17.9) | 65.3 (10.9) | 52.5 (16.7) |
Median (range) | 52 (4 - 84) | 56 (14 - 95) | 65 (49 - 86) | 54 (4 - 95) |
Tumor Thickness (mm) | ||||
Mean (SD) | 2.24 (2.33) | 4.31 (3.57) | 3.72 (3.63) | 2.92 (2.96) |
Median (range) | 1.36 (0.1 - 12.9) | 3.54 (0.32 - 17.4) | 2.26 (0.16 - 12.6) | 1.94 (0.1 - 17.4) |
The default presentation includes “N =” under the header, but also has percent characters in the table. Considerable gymnastics are required to make this work in latex. For html, we simply replace the newline character (supplied by ‘lab’) with a ‘br’ tag internally.
tablet
gives a flexible way of summarizing tables of
observations. It reacts to numeric columns, factors, and grouping
variables. Display order derives from the order of columns and factor
levels in the data. Result columns can be grouped arbitrarily deep by
supplying extra groups. Column labels and titles are respected.
Rendering is largely the responsibility of kableExtra
and
can be extended. Further customization is possible by manipulating data
after calling tablet() but before calling as_kable(). Powerful results
are possible with very little code.