The jfbr package is meant to test and share miscellaneous functions
for quantitative workflows.
You can install the development version of jfbr from GitHub with:
# install.packages("devtools") # Install devtools if not already installed
devtools::install_github("jbetz-jhu/jfbr")Convenience functions have been added for the table1::table1()
function to make it easier to do different summaries and add hypothesis
testing.
library(jfbr)
library(table1)
#>
#> Attaching package: 'table1'
#> The following objects are masked from 'package:base':
#>
#> units, units<-
library(knitr)The package includes an example dataset jfbr_test which is used in the
examples below:
head(jfbr_test)
#> numbers continuous binary ordered binary_factor categorical two_level_group
#> 1 1 0.7209039 1 3 1. Yes 1 Group 1
#> 2 2 0.8757732 1 2 0. No 4 Group 2
#> 3 3 0.7609823 0 1 1. Yes 1 Group 1
#> 4 4 0.8861246 1 3 1. Yes 3 Group 1
#> 5 5 0.4564810 1 1 0. No 1 Group 2
#> 6 6 0.1663718 1 4 1. Yes 2 Group 1
#> three_level_group
#> 1 Group 1
#> 2 Group 1
#> 3 Group 3
#> 4 Group 3
#> 5 Group 3
#> 6 Group 1The default for table1::table1() is to produce Mean (SD) and Median
[Min, Max] for numeric values, and NA is treated as a level of a
factor for categorical variables.
Note: saving the result of table1::table1() and using
knitr::kable()is only necessary when HTML output is not possible
(e.g. output: github_document in R Markdown)
# table1() defaults
my_table <-
table1(
x = ~ continuous + numbers + ordered + binary_factor + categorical |
two_level_group,
data = jfbr_test
)
kable(my_table)| Group 1 | Group 2 | Overall | |
|---|---|---|---|
| (N=97) | (N=103) | (N=200) | |
| continuous | |||
| Mean (SD) | 0.525 (0.297) | 0.560 (0.291) | 0.543 (0.294) |
| Median [Min, Max] | 0.510 [0.00114, 0.993] | 0.601 [0.00599, 0.990] | 0.588 [0.00114, 0.993] |
| Missing | 0 (0%) | 1 (1.0%) | 1 (0.5%) |
| numbers | |||
| Mean (SD) | 100 (57.0) | 101 (59.0) | 101 (57.9) |
| Median [Min, Max] | 100 [1.00, 200] | 104 [2.00, 197] | 101 [1.00, 200] |
| Missing | 0 (0%) | 1 (1.0%) | 1 (0.5%) |
| ordered | |||
| 1 | 29 (29.9%) | 34 (33.0%) | 63 (31.5%) |
| 2 | 15 (15.5%) | 25 (24.3%) | 40 (20.0%) |
| 3 | 28 (28.9%) | 20 (19.4%) | 48 (24.0%) |
| 4 | 24 (24.7%) | 24 (23.3%) | 48 (24.0%) |
| Missing | 1 (1.0%) | 0 (0%) | 1 (0.5%) |
| binary_factor | |||
| 0. No | 46 (47.4%) | 47 (45.6%) | 93 (46.5%) |
| 1. Yes | 51 (52.6%) | 55 (53.4%) | 106 (53.0%) |
| Missing | 0 (0%) | 1 (1.0%) | 1 (0.5%) |
| categorical | |||
| 1 | 23 (23.7%) | 24 (23.3%) | 47 (23.5%) |
| 2 | 26 (26.8%) | 16 (15.5%) | 42 (21.0%) |
| 3 | 25 (25.8%) | 30 (29.1%) | 55 (27.5%) |
| 4 | 23 (23.7%) | 32 (31.1%) | 55 (27.5%) |
| Missing | 0 (0%) | 1 (1.0%) | 1 (0.5%) |
Using the argument render.continuous = table1_numeric adds Median
[IQR] and [Max, Min], while
render.categorical = table1_categorical summarizes the observed
frequencies and tabulates the proportion of missing values separately:
my_table <-
table1(
x = ~ continuous + numbers | two_level_group,
data = jfbr_test,
render.continuous = table1_numeric,
render.categorical = table1_categorical
)
kable(my_table)| Group 1 | Group 2 | Overall | |
|---|---|---|---|
| (N=97) | (N=103) | (N=200) | |
| continuous | |||
| Mean (SD) | 0.53 (0.3) | 0.56 (0.29) | 0.54 (0.29) |
| Median [IQR] | 0.51 (0.3) | 0.6 (0.29) | 0.59 (0.29) |
| [Min, Max] | [0, 0.99] | [0.01, 0.99] | [0, 0.99] |
| Complete (N%) | 97 (100%) | 102 (99.03%) | 199 (99.5%) |
| Missing | 0 (0%) | 1 (1.0%) | 1 (0.5%) |
| numbers | |||
| Mean (SD) | 100.05 (57.01) | 101.42 (59.03) | 100.75 (57.91) |
| Median [IQR] | 100 (57.01) | 103.5 (59.03) | 101 (57.91) |
| [Min, Max] | [1, 200] | [2, 197] | [1, 200] |
| Complete (N%) | 97 (100%) | 102 (99.03%) | 199 (99.5%) |
| Missing | 0 (0%) | 1 (1.0%) | 1 (0.5%) |
The arguments mean_sd, median_iqr, and range control which
summaries are computed. Quantiles can be added optionally with the
quantiles argment:
# Only Mean/SD
my_table <-
table1(
x = ~ continuous + numbers | two_level_group,
data = jfbr_test,
render.continuous =
function(x)
table1_numeric(
x = x,
mean_sd = TRUE, median_iqr = FALSE, range = FALSE,
quantiles = NULL
)
)
kable(my_table)| Group 1 | Group 2 | Overall | |
|---|---|---|---|
| (N=97) | (N=103) | (N=200) | |
| continuous | |||
| Mean (SD) | 0.53 (0.3) | 0.56 (0.29) | 0.54 (0.29) |
| Complete (N%) | 97 (100%) | 102 (99.03%) | 199 (99.5%) |
| Missing | 0 (0%) | 1 (1.0%) | 1 (0.5%) |
| numbers | |||
| Mean (SD) | 100.05 (57.01) | 101.42 (59.03) | 100.75 (57.91) |
| Complete (N%) | 97 (100%) | 102 (99.03%) | 199 (99.5%) |
| Missing | 0 (0%) | 1 (1.0%) | 1 (0.5%) |
# Only Mean/SD, 5% and 95% Quantiles
my_table <-
table1(
x = ~ continuous + numbers | two_level_group,
data = jfbr_test,
render.continuous =
function(x)
table1_numeric(
x = x,
mean_sd = TRUE, median_iqr = FALSE, range = FALSE,
quantiles = c(0.05, 0.95)
)
)
kable(my_table)| Group 1 | Group 2 | Overall | |
|---|---|---|---|
| (N=97) | (N=103) | (N=200) | |
| continuous | |||
| Mean (SD) | 0.53 (0.3) | 0.56 (0.29) | 0.54 (0.29) |
| 5%, 95% | 0.03, 0.97 | 0.09, 0.96 | 0.05, 0.97 |
| Complete (N%) | 97 (100%) | 102 (99.03%) | 199 (99.5%) |
| Missing | 0 (0%) | 1 (1.0%) | 1 (0.5%) |
| numbers | |||
| Mean (SD) | 100.05 (57.01) | 101.42 (59.03) | 100.75 (57.91) |
| 5%, 95% | 10.2, 189 | 12.05, 189.95 | 10.9, 190.1 |
| Complete (N%) | 97 (100%) | 102 (99.03%) | 199 (99.5%) |
| Missing | 0 (0%) | 1 (1.0%) | 1 (0.5%) |
Hypothesis tests can be added to table1 using the extra.col
argument: there is a worked example of including t.test and
chisq.test in the table1
documentation.
The table1_pvalue function is a convenience function that allows users
to supply their own tests to be computed in table1. The defaults
include t.test and ANOVA omnibus test (via a lm and anova wrapper)
for continuous variables, and chisq.test for categorical variables:
my_table <-
table1::table1(
x = ~ numbers + continuous + binary + ordered + binary_factor +
categorical | two_level_group,
data = jfbr_test,
overall = FALSE,
extra.col =
list("p-value" = table1_pvalue)
)
kable(my_table)| Group 1 | Group 2 | p-value | |
|---|---|---|---|
| (N=97) | (N=103) | ||
| numbers | |||
| Mean (SD) | 100 (57.0) | 101 (59.0) | 0.868 |
| Median [Min, Max] | 100 [1.00, 200] | 104 [2.00, 197] | |
| Missing | 0 (0%) | 1 (1.0%) | |
| continuous | |||
| Mean (SD) | 0.525 (0.297) | 0.560 (0.291) | 0.407 |
| Median [Min, Max] | 0.510 [0.00114, 0.993] | 0.601 [0.00599, 0.990] | |
| Missing | 0 (0%) | 1 (1.0%) | |
| binary | |||
| Mean (SD) | 0.546 (0.500) | 0.578 (0.496) | 0.651 |
| Median [Min, Max] | 1.00 [0, 1.00] | 1.00 [0, 1.00] | |
| Missing | 0 (0%) | 1 (1.0%) | |
| ordered | |||
| 1 | 29 (29.9%) | 34 (33.0%) | 0.263 |
| 2 | 15 (15.5%) | 25 (24.3%) | |
| 3 | 28 (28.9%) | 20 (19.4%) | |
| 4 | 24 (24.7%) | 24 (23.3%) | |
| Missing | 1 (1.0%) | 0 (0%) | |
| binary_factor | |||
| 0. No | 46 (47.4%) | 47 (45.6%) | 0.962 |
| 1. Yes | 51 (52.6%) | 55 (53.4%) | |
| Missing | 0 (0%) | 1 (1.0%) | |
| categorical | |||
| 1 | 23 (23.7%) | 24 (23.3%) | 0.24 |
| 2 | 26 (26.8%) | 16 (15.5%) | |
| 3 | 25 (25.8%) | 30 (29.1%) | |
| 4 | 23 (23.7%) | 32 (31.1%) | |
| Missing | 0 (0%) | 1 (1.0%) |
my_table <-
table1::table1(
x = ~ numbers + continuous + binary + ordered + binary_factor +
categorical | three_level_group,
data = jfbr_test,
overall = FALSE,
extra.col =
list("p-value" = table1_pvalue)
)
kable(my_table)| Group 1 | Group 2 | Group 3 | p-value | |
|---|---|---|---|---|
| (N=67) | (N=70) | (N=63) | ||
| numbers | ||||
| Mean (SD) | 103 (61.3) | 97.7 (52.6) | 101 (60.6) | 0.85 |
| Median [Min, Max] | 107 [1.00, 197] | 99.5 [7.00, 185] | 98.0 [3.00, 200] | |
| Missing | 0 (0%) | 0 (0%) | 1 (1.6%) | |
| continuous | ||||
| Mean (SD) | 0.521 (0.321) | 0.555 (0.282) | 0.553 (0.280) | 0.756 |
| Median [Min, Max] | 0.509 [0.00693, 0.990] | 0.600 [0.00114, 0.970] | 0.609 [0.00865, 0.993] | |
| Missing | 0 (0%) | 1 (1.4%) | 0 (0%) | |
| binary | ||||
| Mean (SD) | 0.552 (0.501) | 0.507 (0.504) | 0.635 (0.485) | 0.332 |
| Median [Min, Max] | 1.00 [0, 1.00] | 1.00 [0, 1.00] | 1.00 [0, 1.00] | |
| Missing | 0 (0%) | 1 (1.4%) | 0 (0%) | |
| ordered | ||||
| 1 | 13 (19.4%) | 29 (41.4%) | 21 (33.3%) | 0.00748 |
| 2 | 21 (31.3%) | 11 (15.7%) | 8 (12.7%) | |
| 3 | 14 (20.9%) | 20 (28.6%) | 14 (22.2%) | |
| 4 | 18 (26.9%) | 10 (14.3%) | 20 (31.7%) | |
| Missing | 1 (1.5%) | 0 (0%) | 0 (0%) | |
| binary_factor | ||||
| 0. No | 26 (38.8%) | 40 (57.1%) | 27 (42.9%) | 0.0825 |
| 1. Yes | 41 (61.2%) | 30 (42.9%) | 35 (55.6%) | |
| Missing | 0 (0%) | 0 (0%) | 1 (1.6%) | |
| categorical | ||||
| 1 | 18 (26.9%) | 15 (21.4%) | 14 (22.2%) | 0.364 |
| 2 | 13 (19.4%) | 12 (17.1%) | 17 (27.0%) | |
| 3 | 22 (32.8%) | 18 (25.7%) | 15 (23.8%) | |
| 4 | 13 (19.4%) | 25 (35.7%) | 17 (27.0%) | |
| Missing | 1 (1.5%) | 0 (0%) | 0 (0%) |
Any function that returns an element p.value can be passed as an
argument, allowing users to customize which tests are performed:
my_table <-
table1::table1(
x = ~ numbers + continuous + binary + ordered + binary_factor +
categorical | two_level_group,
data = jfbr_test,
overall = FALSE,
extra.col =
list("p-value" =
function(x, value) table1_pvalue(
x = x,
variable = variable,
test_numeric_2_levels = wilcox.test,
test_numeric_more_than_2_levels = kruskal.test,
test_categorical_2_levels = fisher.test,
test_categorical_more_than_2_levels = fisher.test
)
)
)
kable(my_table)| Group 1 | Group 2 | p-value | |
|---|---|---|---|
| (N=97) | (N=103) | ||
| numbers | |||
| Mean (SD) | 100 (57.0) | 101 (59.0) | 0.866 |
| Median [Min, Max] | 100 [1.00, 200] | 104 [2.00, 197] | |
| Missing | 0 (0%) | 1 (1.0%) | |
| continuous | |||
| Mean (SD) | 0.525 (0.297) | 0.560 (0.291) | 0.377 |
| Median [Min, Max] | 0.510 [0.00114, 0.993] | 0.601 [0.00599, 0.990] | |
| Missing | 0 (0%) | 1 (1.0%) | |
| binary | |||
| Mean (SD) | 0.546 (0.500) | 0.578 (0.496) | 0.651 |
| Median [Min, Max] | 1.00 [0, 1.00] | 1.00 [0, 1.00] | |
| Missing | 0 (0%) | 1 (1.0%) | |
| ordered | |||
| 1 | 29 (29.9%) | 34 (33.0%) | 0.266 |
| 2 | 15 (15.5%) | 25 (24.3%) | |
| 3 | 28 (28.9%) | 20 (19.4%) | |
| 4 | 24 (24.7%) | 24 (23.3%) | |
| Missing | 1 (1.0%) | 0 (0%) | |
| binary_factor | |||
| 0. No | 46 (47.4%) | 47 (45.6%) | 0.888 |
| 1. Yes | 51 (52.6%) | 55 (53.4%) | |
| Missing | 0 (0%) | 1 (1.0%) | |
| categorical | |||
| 1 | 23 (23.7%) | 24 (23.3%) | 0.241 |
| 2 | 26 (26.8%) | 16 (15.5%) | |
| 3 | 25 (25.8%) | 30 (29.1%) | |
| 4 | 23 (23.7%) | 32 (31.1%) | |
| Missing | 0 (0%) | 1 (1.0%) |