univ: Univariate Table

Description

Similar to the QHS SAS macro, provides a numerical simple summary of numerical variables.

Usage

univ(data, vars, byVar, alpha=0.05,
     test=c("t.test", "aov"),
     test.args=NULL)

Arguments

data

A ccf.df or data.frame containing the variables in vars and byVar.

vars

Character vector naming the variables in data to be summarized.

byVar

A categorical variables giving groups by which statistics should be computed.

alpha

significance level for determining the confidence limits.

test

The test to use for comparing between groups. Currently limited to t-test and ANOVA (parametric tests).

test.args

a list of additional arguments to pass to the function in test.

code

conttable

Details

Statistics available in univ, and the calls to get them are:

{number of non-missing values.} missing{number of missing values} mean{arithmetic mean} median{median value} sd{standard deviation} lcl{lower confidence limit} ucl{upper confidence limit} min{minimum value} max{maximum value} p25{25th percentile} p75{75th percentile} cv{coefficient of variation}

Examples

Run this code

#Read in the delivery dataset from the CCFmisc library
require(Hmisc)

data(Delivery)

#use Hmisc library to labeling variables in univariate tables
label(Delivery$maternal.age) <- "Maternal Age"
label(Delivery$ga.weeks) <- "Gestation weeks"
label(Delivery$wt.gram) <- "Weight (g)"


#a univariate table of the variables maternal age,
#ga.weeks and wt.grams.  The object resulting
#from univ() can be used in other functions to create html or
#LaTeX tables.

uni <- univ(Delivery,
            vars=c("maternal.age", "ga.weeks", "wt.gram"))

#a univariate table of the variables maternal age,
#ga.weeks and wt.grams by delivery.type.  The object resulting
#from univ() can be used in other functions to create html or
#LaTeX tables.

deliv.uni <- univ(Delivery,
	vars=c("maternal.age", "ga.weeks", "wt.gram"),
	byVar="delivery.type")

#if you want to take advantage of the confidence interval
#output from univ() different alpha levels can be set
#by the alpha= argument.

deliv_99.uni <- univ(Delivery,
	vars=c("maternal.age", "ga.weeks", "wt.gram"),
	byVar="delivery.type",
	alpha=0.01)