Report results from nonparametric tests

This repository provides the function report_nonparametrics to automatically generate the results from the most common nonparametric tests, like Pearson's chi-squared test, Kruskal-Wallis or Fisher's Z. If you have questions, feedback, or further ideas, feel free to contact me.

Installation

The package is not yet published on CRAN. Nevertheless, you can install this package from Github. Type in the following lines of code:

install.packages("remotes")

remotes::install_github("huelemeier/report-nonparametrics")

Load the package every time you start R:

library(reportnonparametrics)

The report_nonparametrics() is compatible with:

chisq.test() # Pearson's chi-squared test
kruskal.test() # Kruskal-Wallis rank sum test
friedman.test() # Friedman rank sum test
fisher.test() # Fisher's exact test

Examples

## Pearson's chi-squared test
# Data preparation
iris$size <- ifelse(iris$Sepal.Length < median(iris$Sepal.Length),
  "small", "big")
  
# report findings
report_non_parametrics(chisq.test(table(iris$Sepal.Length, iris$size))
We performed Pearson's Chi-squared test of independence to assess the relationship between table(iris$Sepal.Length, iris$size). At the 5% significance level, the data provide evidence to conclude that there is a significant association between the two variables, (X2(34) = 150, p = .000)

report_non_parametrics(chisq.test(table(iris$Sepal.Length, iris$size), simulate.p.value = TRUE, B = 429))
We performed Pearson's Chi-squared test with simulated p-value (based on 429 replicates) to assess the relationship between table(iris$Sepal.Length, iris$size). At the 5% significance level, the data provide evidence to conclude that there is a significant association between the two variables, (X2(NA) = 150, p = .002)


## Kruskal-Wallis test 
report_nonparametrics(kruskal.test(iris$Petal.Length,iris$Species))
We performed Kruskal-Wallis rank sum test to assess the median difference between iris$Petal.Length and iris$Species. We found one or more of the groups has a different median and, thus, comes from a different distribution. In other words, at the 5% significance level, we conclude that at least one of the variables performs differently than the others, (H(2) = 130.41, p = .000). 
  
  
## Friedman rank sum test
# Data preparation
data <- data.frame(person = rep(1:5, each=4),
                   drug = rep(c(1, 2, 3, 4), times=5),
                   score = c(30, 28, 16, 34, 14, 18, 10, 22, 24, 20,
                             18, 30, 38, 34, 20, 44, 26, 28, 14, 30))
                             
report_nonparametrics(friedman.test(data$score, data$drug, data$person))
A non-parametric Friedman rank sum test among repeated measures of data$score depending on the grouping and block variables data$drug and data$person was conducted. The test rendered a significant Chi-square value suggesting the effect differs between groups (X2(3) = 13.56, p = .004).  

## Fisher's exact test
report_nonparametrics(fisher.test(table(iris$Sepal.Length, iris$size), alternative = "less"))
The Fisher's Exact Test for Count Data was applied to determine if there was a significant association between table(iris$Sepal.Length, iris$size). The results suggest the effect is statistically significant, thus confirming a relation between the two variables (p = .000). We applied a one-sided test assuming a negative association.