summarize: Produce Summary Tables for Data Sets
In hofnerb/papeR: A Toolbox for Writing Pretty Papers and Reports

Description Usage Arguments Value Author(s) See Also Examples

The function produces summary tables for factors and continuous variables. The obtained tables can be used directly in R, with LaTeX and HTML (by using the xtable function) or Markdown (e.g. by using the function kable).

summarize(data, type = c("numeric", "factor"),
    variables = names(data), variable.labels = labels, labels = NULL,
    group = NULL, test = !is.null(group), colnames = NULL,
    digits = NULL, digits.pval = 3, smallest.pval = 0.001,
    sep = NULL, sanitize = TRUE, drop = TRUE,
    show.NAs = any(is.na(data[, variables])), ...)

`data`	data set to be used.
`type`	print summary table for either `numeric` or `factor` variables.
`variables`	character vector defining variables that should be included in the table. Per default, all numeric or factor variables of `data` are used, depending on `type`.
`variable.labels, labels`	labels for the variables. If `variable.labels = NULL` (default) `variables` is used as label. If `variable.labels = TRUE`, `labels(data, which = variables)` is used as labels. Instead of `variable.labels` one can also use `labels`.
`group`	character specifying a grouping factor. Per default no grouping is applied.
`test`	logical or character string. If a `group` is given, this argument determines whether a test for group differences is computed. For details see `summarize_numeric` and `summarize_factor`.
`colnames`	a vector of character strings of appropriate length. The vector supplies alternative column names for the resulting table. If `NULL` default names are used.
`digits`	number of digits to round to. For defaults see `summarize_numeric` and `summarize_factor`.
`digits.pval`	number of significant digits used for p-values.
`smallest.pval`	determines the smallest p-value to be printed exactly. Smaller values are given as “< smallest.pval”. This argument is passed to the `eps` argument of `format.pval`. See there for details.
`sep`	logical. Determines whether separators (lines) should be added after each variable. For defaults see `summarize_numeric` and `summarize_factor`.
`sanitize`	logical (default: `TRUE`) or a sanitizing function used to clean the input in order to be useable in LaTeX environments. Per default `toLatex.character` is used.
`drop`	logical (default: `TRUE`). Determines whether variables, which contain only missing values are dropped from the table.
`show.NAs`	logical. Determines if NAs are displayed. Per default, `show.NAs` is `TRUE` if there are any missings in the variables to be displayed (and `FALSE` if not). For details see `summarize_numeric` and `summarize_factor`.
`...`	additional arguments for `summarize_numeric` and `summarize_factor`. See there for details.

A special data.frame with additional class summary containing the computed statistics is returned from function summarize. Addtional attributes required for the xtable.summary or print.xtable.summary function are contained as attributes. These are extracted using the function get_option.

Benjamin Hofner

For details see summarize_numeric and summarize_factor.

Conversion to LaTeX tables can be done using xtable.summary and print.xtable.summary.

get_option

if (require("nlme")) {
    ## Use dataset Orthodont
    data(Orthodont, package = "nlme")

    ## Get summary for continuous variables
    (tab1 <- summarize(Orthodont, type = "numeric"))

    ## Change statistics to display
    summarize(Orthodont, quantiles = FALSE, type = "numeric")
    summarize(Orthodont, quantiles = FALSE, count = FALSE, type = "numeric")
    summarize(Orthodont, mean_sd = FALSE, type = "numeric")

    ## Get summary for categorical variables
    (tab2 <- summarize(Orthodont, type = "fac"))

    ## use fraction instead of percentage
    summarize(Orthodont, percent = FALSE, type = "fac")

    ## Using the tables with Markdown
    if (require("knitr")) {
        kable(tab1)
        kable(tab2)
    }

    ## Using the tables with LaTeX
    if (require("xtable")) {
        xtable(tab1)
        ## grouped table
        xtable(summarize(Orthodont, group = "Sex"))
        xtable(tab2)
    }
}