In joon-e/jutilities: jutilities: A Collection of Convenience Functions

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  out.width = "100%",
  fig.path = "man/figures/"
)

jutilities

This a collection of convenience functions for data handling, plotting, and analysis within the tidyverse syntax. Most of these functions are wrappers for tasks I perform regularly (e. g., describe variables, add mean indices) and wanted to integrate them into %>% pipes with just one line.

This is all very much work in progress.

Installation

Install from GitHub:

install.packages("devtools")
devtools::install_github("joon-e/jutilities")

Instructions

jutilities are to be used with the tidyverse. Most functions will return a tibble and can thus be easily integrated into %>% pipes.

library(tidyverse)
library(jutilities)

Describe continous variables

describe() computes several measures of central tendency and variability for all specified variables:

diamonds %>% 
  describe(x, y, z)

If no variables are specified, all numeric variables are described:

diamonds %>% 
  describe()

describe_groups() outputs the same statistics for one variable, grouped by one or more grouping variables.

diamonds %>% 
  describe_groups(price, cut, color)

Describe categorical variables

cat_ftable() outputs a frequency table including relative, valid, and cumulative frequencies for one categorical variable.

diamonds %>% 
  cat_ftable(cut)

cat_xtable outputs contigency tables for one column variable and one or more row variables:

diamonds %>% 
  cat_xtable(cut, color, clarity)

Setting the argument percentages = TRUE changes the output to relative frequencies:

diamonds %>% 
  cat_xtable(cut, color, percentages = TRUE)

A Chi² test can be optionally computed by setting the argument chisq = TRUE. Test results will be displayed in a console message:

diamonds %>% 
  cat_xtable(cut, color, chisq = TRUE)

Modify data

add_index() adds a rowwise mean index columns of the specified variables to the dataset. The second argument (first argument if used in a pipe) should be the name of the index column:

diamonds %>%
  add_index(meanxyz, x, y, z)

Set the argument type = "sum" to create a sum index instead:

diamonds %>% 
  add_index(sumxyz, x, y, z, type = "sum")

add_label adds a text label column of a numeric variable to the dataset, for example to be used as labels in plots.

diamonds %>% 
  add_label(labelz, z)

By default, add_label rounds to two decimal places. You can change this by setting the decimal.places argument:

diamonds %>% 
  add_label(labelz, z, decimal.places = 0)

Analyze data

Correlations

correlate() computes correlations coefficients and p-values for all combinations of the specified variables. By default, Pearson correlation coefficients are computed. Set the argument type = "spearman" to compute Spearman ranked correlations coefficicents instead.

diamonds %>% 
  correlate(x, y, z)

T-tests

ttest computes t-Tests for one grouping variable and one or more test variables. Output statistics include descriptives, t-values, degrees of freedom, p-values, and Cohen's d:

diamonds %>% 
  filter(cut == "Ideal" | cut == "Fair") %>% 
  droplevels() %>% 
  ttest(cut, price, x, y, z)

joon-e/jutilities documentation built on May 28, 2019, 7:34 p.m.