tableone: Tableone summarizes a selection of the data set as would be...
In dewittpe/qwraps: Quick Wraps

Description Usage Arguments Details Value Author(s) Examples

Summarize variables by a categorical variable within a given data.frame. Overall and by results are presented along with simiple hypothesis tests.

tableone(vars, by = NULL, data = NULL, complete = TRUE, margin = 2,
  overall = TRUE, stat.con.1 = mean, stat.con.2 = sd,
  tests = c("chisq.test", "t.test"), fisher = TRUE,
  digits = getOption("qwraps.digits", 3), big.mark = ",")

## S3 method for class 'tableone'
print(tab1, file = "", title = "",
  ctable = getOption("qwraps.tableone.ctable", TRUE), cgroup, n.cgroup,
  rgroup, n.rgroup, rowname, col.just, ...)

`vars`	characer vector of the names of the variables in data to summarize.
`by`	character name of the variable in data to sumamrize vars by
`data`	the data.frame
`complete`	logical reported for the complete data records only? If TRUE, the dataset will be subseted via `na.omit(data[, c(vars, by)])`, if FALSE, then all avaiable data records will be used.
`margin`	default to report column percentages for categorical variables, can be set to 1 for row percentages or 1:2 for overall percentages.
`overall`	logical if TRUE include a column with the summary statistics for the whole sample in addition to the summary of the `vars` `by`
`stat.con.1`	function to summarize continuous variables with, default is the `mean`
`stat.con.2`	second function to sumamry continuous variables with, adefault is `sd`
`tests`	a vector of length two with the statistical tests to report for 1) categorical variables and 2) continuous variables. If set to NULL, then the tests and associated p-values are omitted from the output. Defaults are for the `chisq.test` and the `t.test`. If the value is set to NULL the tests will be omitted from the output. Only `chisq.test` and `fisher.test` are allowed for categorical variables at this time. Regardless of the test function specified, the function should return a list value named `p.value`. The second test can be either a t.test or a aov. more tests will be allowed in later versions.
`fisher`	logicial if TRUE and tests[1] == "chisq.test" then if `any(chisq.test()$expected) < 5` the test is redone using a `fisher.exact`. This option is ignored if `tests[1] == "fisher.test"`.
`big.mark`	Default is a comma to separate, i.e., 1000 would be reported as 1,000.
`digits`	number of digits to round stat.con.1 and stat.con.2
`pdigits`	number of digits to report p-values to.
`show.equal.sign`

EXPLAIN THE DETAILS MORE, FACTORS AND NUMERS ARE IMPORANT, WILL TRANSFORM CHARACTERS TO FACTORS

a list with the elements intended to be used with Hmisc::latex

`tab`	a string of formated numbers. Non-numbers are empty strings. If the p-value is less than 10^(-pdigits) the return will be '< 10^(-pdigits)
`cgrp`	a character vector for the column groups, the levels of `by`
`ncgrp`	a numeric vector for n.cgroup
`rgrp`	a character vector for the row groups
`rcgrp`	a numeric vector for n.cgroup
`rwnms`	a character vector of row names

Peter DeWitt

data(diamonds, package = "ggplot2")

## simple examples for the just getting the matrix for table one
## this could be used in xtable, Hmisc::latex, and likely other functions
tableone("cut", "color", diamonds[1:50, ])$tab
tableone("price", "color", diamonds[1:50, ],
         tests = c("chisq.test", #' "aov"))$tab
tableone(c("cut", "price"), "color", diamonds[1:50, ],
         tests = c("chisq.test", "aov"))$tab

## A LaTeX table via Hmisc::latex

tab1 <- tableone(c("cut", "price"), "color", diamonds[1:50, ],
                 tests = c("chisq.test", "aov"))
Hmisc::latex(tab1$tab,
             file     = "",
             title    = "Example from qwraps",
             ctable   = TRUE,
             cgroup   = tab1$cgrp,
             n.cgroup = tab1$ncgrp,
             colhead  = NULL,
             rgroup   = tab1$rgrp,
             n.rgroup = tab1$nrgrp,
             rowname  = tab1$rwnm,
             caption  = "Example Table 1 from the qwraps with Hmisc.",
             label    = "tab:tableone",
             col.just = rep("r", ncol(tab1$tab)))

## A better and easier way to get the table into a LaTeX file
print(tab1)