report_continuous: Report summary table of means±sd or medians±IQR...

report_continuousR Documentation

Report summary table of means±sd or medians±IQR (Interquartile Range)

Description

report_medianIQR builds a summary table with the overall mean and sd or medians and quartiles of one or more variables depending on whether the variable is normal or not. A grouping variable can also be specified, in which case means and sd (or medians and quartiles) are also calculated for each group. When means and sd are reported, the p-value of an anova test for testing the significance of the differences between the means is shown. When the reported values are medians and quartiles, the p-value of a Kruskal-Wallis test is displayed to test the null hypothesis that the location parameters of the distribution of the variable are the same in each group.

Usage

report_continuous(
  data,
  summary_vars,
  groupVar = NULL,
  digits = 2,
  probs = c(0.25, 0.75),
  pvdigits = 4,
  alpha = 0.05,
  na.rm = TRUE
)

Arguments

data

data frame or tibble which contains the data.

summary_vars

Variable or variables whose summary measures (mean±sd or median±IQR) are to be calculated.

groupVar

Grouping variable.

digits

Number of decimal digits for the results.

probs

Pair of quantiles to be computed around the median. Default are q25 and q75.

pvdigits

Number of decimal digits for the p-value of Kruskal test.

alpha

Decision on computing mean±sd or median±IQR is made depending on the result of a Shapiro Wilk test of normality. The value of alpha is the significance level for that test.

na.rm

Should NA values be removed (possible values are TRUE or FALSE)

roundFrom

If median is greater than this value, median and IQR are rounded to zero decimals even if digits>0.

Value

A table with the overall median and quartiles of the variables and, if a grouping variable is specified, the medians and quartiles by group and the p-value of the kruskal test for comparing location parameters.

Examples

df <- data.frame(x=rnorm(100,10,3),y=rnorm(100,50,8), z=runif(100,20,30),g=sample(c("Yes","No"),100,replace=TRUE))
df %>%
report_continuous(c(x,y,z))  # Only overall summary of variables x, y and z
df %>%
report_continuous(c(x,y,z), groupVar=g, digits=1)

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