Skip to contents

lsr

lsr is a companion package to the textbook Learning Statistics with R. It provides beginner-friendly wrappers for common statistical procedures — t-tests, chi-square tests, effect sizes, correlation matrices, and basic data manipulation — with output designed to be readable by students encountering statistics for the first time.

Example

Here is an independent-samples t-test comparing extra sleep between two drug groups in the built-in sleep dataset:

library(lsr)
independentSamplesTTest(formula = extra ~ group, data = sleep)
#> 
#>    Welch's independent samples t-test 
#> 
#> Outcome variable:   extra 
#> Grouping variable:  group 
#> 
#> Descriptive statistics: 
#>                 1     2
#>    mean     0.750 2.330
#>    std dev. 1.789 2.002
#> 
#> Hypotheses: 
#>    null:        population means equal for both groups
#>    alternative: different population means in each group
#> 
#> Test results: 
#>    t-statistic:  -1.861 
#>    degrees of freedom:  17.776 
#>    p-value:  0.079 
#> 
#> Other information: 
#>    two-sided 95% confidence interval:  [-3.365, 0.205] 
#>    estimated effect size (Cohen's d):  0.832

Compared to base R’s t.test(), the output labels every component in plain English and automatically reports Cohen’s d alongside the test result.

Where to go next

  • Guided overview — a hands-on introduction for students and beginners, working through descriptive statistics, reshaping data, and hypothesis testing with a single example dataset.
  • Critical commentary — an honest account of the package’s limitations and pointers to better tools for users who have outgrown it.
  • Reference — documentation for all 29 functions, organised by topic.

Installation

Install the released version from CRAN:

Or the development version from GitHub:

# install.packages("devtools")
devtools::install_github("djnavarro/lsr")