MetricsBySplit produces a data.frame describing one or many metric variables by a grouping factor SPLIT

MetricsBySplit produces a data.frame describing one or several metric variables with various statistics SPLIT is a group VARNAME which may be left out. Tests may be preformed on Mean (t.test, oneway.test) or Median (wilcox.test,kruskal.test) MetricsBySplit is a wrapper function calling MetricBySplit for each COLUMN, rbinding the results.

MetricsBySplit(DFR, COLUMNS, SPLIT = NULL, LONGNAMES = NULL,
  PVALUE = F, TEST = NULL, ALL = T, SEL = NULL, DIGITS = 2,
  SHOW_N = F)

Arguments

DFR	is a dataframe
COLUMNS	colnames in DFR to be described (at least one)
SPLIT	=NULL split factor specified by a colname in DFR or as vector of fitting length. If NULL an ungrouped description is returned.
LONGNAMES	=NULL alternative names for the COLUMNS
PVALUE	=F If TRUE, tests on mean and Median are performed and added as last column
TEST	=NULL Choose TEST from c("Mean","Median") oce for all or for each COLUMN. If not specified set to both i.e. c("Mean","Median")
ALL	=T If TRUE, also overall statistics
SEL	=NULL if specified selects statistics from c("Nvalid", "Mean", "Sd", "Median", "Mad", "Perc25", "Perc75", "Min", "Max") [once for all columns]
DIGITS	=2 Number of digits for statistics
SHOW_N	=T Show N=length(METRIC) top left

Value

a data.frame OF CHARACTERS with statistics for each COLUMN by the group variableSPLIT

Details

MetricsBySplit

See also

CrossTabs

Examples

if (FALSE) {
DFR<-data.frame( Metric1=c(rnorm(57),rnorm(43,2,2)),
                 Metric2=c(rnorm(57,1,2),rnorm(43,20,5)),
                 Split=c(rep("Arm A",57),rep("Arm B",43)) )
MetricsBySplit(DFR,COLUMNS=c("Metric1","Metric2"),SPLIT="Split")
MetricsBySplit(DFR,COLUMNS=c("Metric1","Metric2"),SPLIT=DFR$Split,ALL=F)
MetricsBySplit(DFR,COLUMNS=c("Metric1","Metric2"),SEL=c("Mean","Sd"))
}