Provides Batch Functions and Visualisation for Basic Statistical
Procedures
Description
Designed to streamline data analysis and statistical testing, reducing the length of R
scripts while generating well-formatted outputs in 'pdf', 'Microsoft Word', and 'Microsoft Excel'
formats. In essence, the package contains functions which are sophisticated wrappers around
existing R functions that are called by using 'f_' (user f_riendly) prefix followed by the normal
function name. This first version of the 'rfriend' package focuses primarily on data exploration,
including tools for creating summary tables, f_summary(), performing data transformations,
f_boxcox() in part based on 'MASS/boxcox' and 'rcompanion', and f_bestNormalize()
which wraps and extends functionality from the 'bestNormalize' package. Furthermore, 'rfriend'
can automatically (or on request) generate visualizations such as boxplots, f_boxplot(),
QQ-plots, f_qqnorm(), histograms f_hist(), and density plots. Additionally, the package includes
four statistical test functions: f_aov(), f_kruskal_test(), f_glm(), f_chisq_test for sequential
testing and visualisation of the 'stats' functions: aov(), kruskal.test(), glm() and chisq.test.
These functions support testing multiple response variables and predictors, while also handling
assumption checks, data transformations, and post hoc tests. Post hoc results are automatically
summarized in a table using the compact letter display (cld) format for easy interpretation. The
package also provides a function to do model comparison, f_model_comparison(), and several
utility functions to simplify common R tasks. For example, f_clear() clears the workspace and
restarts R with a single command; f_setwd() sets the working directory to match the directory
of the current script; f_theme() quickly changes 'RStudio' themes; and f_factors() converts
multiple columns of a data frame to factors, and much more. If you encounter any issues or have
feature requests, please feel free to contact me via email.