report_nPct: Frequency table for categorical variables

In angeloSdP/reportingTools: Tools for building report tables containing overall and per group summary statistics

Frequency table for categorical variables

Description

report_nPct builds a summary table with the frequency and percentage of the different values of one or more categorical variables. If a grouping variable is specified, absolute frequencies and percentages o values are computed in each group, and a chi-squared test is performed to test the association with the grouping variable. If conditions for the validity of the chi-squared test are not met, a Fisher exact test is performed instead.

Usage

report_nPct(
  data,
  summary_vars,
  groupVar = NULL,
  digits = 1,
  pvdigits = 4,
  na.rm = TRUE
)

Arguments

data	data frame or tibble which contains the variables.
summary_vars	Variables whose frequencies and percentages are to be computed
groupVar	Grouping variables.
digits	Number of decimal digits of the result.
pvdigits	Number of decimal digits in the p-value of chi-squared test.
na.rm	Should missing values be included in the table? Percentages are always computed excluding missing values.

Value

A table with the overall frequencies and percentages of the values of the categorical variable and, if a grouping variable is specified, the frequencies and percentages of the values in each group as well as a chi-squared test for association between the variable and the grouping variable.

Examples

df <- data.frame(x=rbinom(90,3,0.6),y=rbinom(90,4,0.8), z=rbinom(90,5,0.5),
g=sample(c("Yes","No"),90,replace=TRUE))
df %>%
report_nPct(c(x,y,z))  # Overall summary of variables x, y and z
df %>%
report_nPct(c(x,y,z), groupVar=g, digits=3)
df$x[sample(1:90,15)]=NA  # Some missing values are included in x
df$y[sample(1:90,8)]=NA  # Some missing values are included in y
df %>% report_nPct(c(x,y,z), groupVar=g, digits=3, na.rm=FALSE)

angeloSdP/reportingTools documentation built on Dec. 25, 2024, 11:19 a.m.