knitr::opts_chunk$set(echo = TRUE)
This vignette is current as of
We will first make a fictitious data set:
df <- data.frame(a = rnorm(1000, 1.5, 2), b = seq(1, 1000, 1), c = c(rep("control", 400), rep("Other", 70), rep("treatment", 500), rep("None", 30)), d = c(sample(1:1000, 900, replace=TRUE), rep(-99, 100)))
There are four functions that we'll demonstrate here:
washer is a great function for quick data cleaning. In situations where there are placeholders, extra levels in a factor, or several values need to be changed to another.
library(dplyr) df <- df %>% mutate(d = washer(d, -99), ## changes the placeholder -99 to NA c = washer(c, "Other", "None", value = "control")) ## changes "Other" and "None" to "Control"
Now that the data is "washed" we can start exploring and reporting.
table1(df, a, b, factor(c), d)
The variables must be numeric or factor. Since we use a special type of evaluation (i.e. Non-Standard Evaluation) we can change the variables in the function (e.g.,
factor(c)). This can be extended to making a whole new variable in the function as well.
table1(df, a, b, d, ifelse(a > 1, 1, 0))
This is just the beginning though. Two powerful things the function can do are shown below:
table1(df, a, b, d, ifelse(a > 1, 1, 0), splitby=~factor(c), test=TRUE)
splitby = ~factor(c) stratifies the means and counts by a factor variable (in this case either control or treatment). When we use this we can also automatically compute tests of significance using
We can also use it intuitively within the pipe (for more about this, see the "Table 1" vignette):
df %>% group_by(c) %>% table1(a, b, d, ifelse(a > 1, 1, 0), test=TRUE)
In this case, we used the
group_by() function from
dplyr (within the
table1() knows to use that as the grouping variable in place of the
If the parametric tests (default) are not appropriate, you can set
param = FALSE.
df %>% group_by(c) %>% table1(a, b, d, ifelse(a > 1, 1, 0), test=TRUE, param=FALSE)
Finally, you can polish it quite a bit using a few other options. For example, you can do the following:
table1(df, a, b, d, ifelse(a > 1, 1, 0), splitby=~factor(c), test=TRUE, var_names = c("A", "B", "D", "New Var"), type = c("simple", "condensed"))
var_names can be used for more complex naming (e.g., with spaces, brackets) that otherwise cannot be used with data frames. Alternatively, for more simple naming, we can name them directly.
table1(df, A = a, B = b, D = d, A2 = ifelse(a > 1, 1, 0), splitby=~factor(c), test=TRUE, type = c("simple", "condensed"))
You can also format the numbers (adding a comma for big numbers such as in 20,000 instead of 20000):
table1(df, a, b, d, ifelse(a > 1, 1, 0), splitby=~factor(c), test=TRUE, var_names = c("A", "B", "D", "New Var"), format_number = TRUE)
The table can be exported directly to a folder in the working directory called "Table1". Using
export, we provide it with a string that will be the name of the CSV containing the formatted table.
table1(df, a, b, d, ifelse(a > 1, 1, 0), splitby=~factor(c), test=TRUE, var_names = c("A", "B", "D", "New Var"), format_number = TRUE, export = "example_table1")
This can also be outputted as a latex, markdown, or pandoc table (matching all the output types of
knitr::kable). Below shows how to do a latex table (not using
kable however, but a built-in function that provides the variable name at the top of the table):
table1(df, a, b, d, "new var" = ifelse(a > 1, 1, 0), splitby = ~factor(c), test = TRUE, output = "latex2")
Last item to show you regarding
table1() is that it can be printed in a simplified and condensed form. This instead of reporting counts and percentages for categorical variables, it reports only percentages and the table has much less white space.
table1(df, a, b, d, "new var" = ifelse(a > 1, 1, 0), splitby = ~factor(c), test = TRUE, type = c("simple", "condensed"))
This function is to create simple, beautiful correlation tables. The syntax is just like
table1() in most respects. Below we include all the numeric variables to see their correlations. Since there are missing values in
d we will use the natural
tableC(df, a, b, d, na.rm = TRUE)
All the adjustments that you can make in
table1() can be done here as well. For example,
tableC(df, "A" = a, "B" = b, "D" = d, na.rm = TRUE, output = "html")
This function is to create simple frequency tables. The syntax is just like
tableC() in most respects, except that it uses only one variable instead of many.
table1() we can use a
splitby argument (or
tableF(df, d, splitby = c)
df %>% group_by(c) %>% tableF(d)
tableX() is a pipe-able two-way version of
table() with a similar syntax to that of the rest of the
df %>% tableX(c, ifelse(d > 500, 1, 0))
By default, it provides the total counts for the rows and columns with flexibility as to what is displayed and where.
The four functions:
washer() add simplicity to cleaning up and understanding your data. Use these pieces of furniture to make your quantitative life a bit easier.
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.