README.md

In thoera/corr: Compute and Visualize Correlations and Cramér's V

corr

Overview

corr aims to provide a consistent interface to compute several types of correlation coefficient and Cramér's V measure of association. It offers an easy way to visualize the results with heatmaps.

corr fits naturally in the tidyverse and is "pipe-friendly".

Installation

To install the package, simply run the following from an R console:

# install.packages("remotes")
remotes::install_github("thoera/corr")

Usage

library("corr")

# compute the correlation
results <- compute_cor(mtcars, method = "pearson")

# view the results 
plot_cor(results, value = TRUE)

# corr is "pipe-friendly"
library("tidyverse")

starwars %>%
  select_if(is.character) %>%
  select(-name) %>%
  filter(complete.cases(.)) %>%
  compute_cor(method = "cramer") %>%
  plot_cor(type = "lower",
           value = TRUE,
           limits_scale = c(0, 1),
           title_legend = "Cramér's V:")

Example

With the mtcars dataset:

thoera/corr documentation built on May 8, 2019, 11:59 p.m.