multGroup: multGroup

In TaylorAndrew/atAnalyze: Methods for Iterative Analytical Reporting

Usage

multGroup(data, PcontinVars = NULL, PcatVars = NULL, NPcontinVars = NULL,
  NPcatVars = NULL, SortVars = NULL, varLabelTable = NULL,
  grouping = NULL, pdec = 3, dec = 2, mu = 0, ChiProbabilities = NULL,
  labels = NULL, percent = "column", NPdescriptives = NULL,
  NPvariance = "range", padjust = F, provideP = T, include = "none",
  verbose = T)

Arguments

data	Input data.frame
PcontinVars	Vector of continous variables names from data that should be tested parametrically
PcatVars	Vector of categorical variables names from data that should be tested parametrically
NPcontinVars	Vector of continous variables names from data that should be tested non-parametrically
NPcatVars	Vector of categorical variables names from data that should be tested non-parametrically
SortVars	Vector of variable names indicating in what order the variables should be presented in the output table
varLabelTable	A varLabelTable object. Converts variable names to variable labels in output table.
grouping	Grouping variable of any number of levels that will be used to stratify each variable
pdec	Number of decimals for p-values
dec	Number of decimals for descriptive statistics
mu	For a one sample t-test (no grouping), the reference value to compare population to. Default of 0.
ChiProbabilities	For a one sample chi-square (no grouping), the underlying probability distribution to compare the population to. If NULL, will compare to an even distribution across all levels.
labels	Labels for the grouping variable
percent	Indicates if the output table should present 'row', 'column', or 'overall' percents for categorical variables.
NPdescriptives	Vector of continuous variable names that should present median and range in the output table.
padjust	If TRUE and grouping has 3 or more levels, only pairwise comparisons that are significant at 0.05 after using a multiple comparisons adjustment are presented
provideP	If TRUE, in the case of no grouping variable, the p-values for the 1-sample tests will be included in the output, otherwise only the descriptive statistics are returned
include	If 'none', no additional output is included. If 'range', range is included. If '95ci' then 95 \item verboseIf TRUE, additional information is printed to the console

A data.frame including all univariate results. multGroup provides summary statistics for continous and categorical variables either overall or stratified by a grouping variable. It also provides parametric or nonparametric tests for each variable.

#df <- data.frame(time = runif(100, 1, 100), # event = sample(c(0,1), 100, replace = T, prob = c(.1, .9)), # x1 = rnorm(100), # x2 = sample(c(1,2), 100, replace = T)) #multGroup(data = df, # NPcontinVars = "time", # grouping = "event", # PcontinVars = "x1", # NPcatVars = "x2", # keepOutput=T, # printOutput = F)

TaylorAndrew/atAnalyze documentation built on May 9, 2019, 4:21 p.m.