The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

kim: A Toolkit for Behavioral Scientists

CRAN checks R build status CodeFactor Total Downloads

This package contains various functions that simplify and expedite analyses of experimental data. Examples include a function that plots sample means of groups in a factorial experimental design, a function that conducts robust regressions with bootstrapped samples, and a function that conducts robust two-way analysis of variance.

Installation

You can install the released version of kim from CRAN with:

install.packages("kim")

You can also install the development version from kim on GitHub with:

install.packages("remotes")
remotes::install_github("jinkim3/kim")

If you run into errors while using the package, try updating the package to the most recent version available on kim on GitHub with:

update_kim()

Example

Here are some examples of using this package.

library(kim)

# (Optional) install all dependencies for all functions in Package 'kim'
install_all_dependencies()

# update the package 'kim', clear the console and environment,
# set up working directory to location of the active document,
# and load the two default packages ('data.table' and 'ggplot2')
start_kim()

# create a scatter plot
scatterplot(data = mtcars, x_var_name = "wt", y_var_name = "mpg")

# get descriptive statistics by group
desc_stats_by_group(
  data = mtcars, var_for_stats = "mpg", grouping_vars = c("vs", "am"))

# plot histograms by group
histogram_by_group(data = mtcars, iv_name = "cyl", dv_name = "mpg")

# plot sample means of groups in a factorial experimental design
plot_group_means(data = mtcars, dv_name = "mpg", iv_name = "gear")

# conduct a two-way ANOVA
factorial_anova_2_way(
  data = mtcars, dv_name = "mpg", iv_1_name = "vs", iv_2_name = "am")

# conduct a multiple regression analysis
multiple_regression(data = mtcars, formula = mpg ~ gear * cyl)

# conduct a robust regression analysis using bootstrapped samples
robust_regression(data = mtcars, formula = mpg ~ cyl * hp)

# conduct a mediation analysis
mediation_analysis(
  data = mtcars, iv_name = "cyl", mediator_name = "disp", dv_name = "mpg")

# conduct a floodlight analysis for a 2 x continuous design
floodlight_2_by_continuous(
  data = mtcars, iv_name = "am", dv_name = "mpg", mod_name = "qsec")

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.