README.md

In condwanaland/summariser: Easy Calculation and Visualisation of Confidence Intervals

summariser

summariser provides simple functions for calculating the most common summary statistics, particularly confidence intervals.

Installation

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

install.packages("summariser")

And the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("condwanaland/summariser")

Using summariser

summariser is designed to fit into the tidyverse ‘piping’ style. Just pass a dataframe, and your measurement variable of interest into summary_stats.

library(summariser)
library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union
iris %>% 
  summary_stats(Sepal.Length)
#>       mean        sd   n         se        ci
#> 1 5.843333 0.8280661 150 0.06761132 0.1325157

If you want to group your dataframe by categorical factors, simply use dplyrs group_by before piping to summary_stats

iris %>%
  group_by(Species) %>% 
  summary_stats(Sepal.Length)
#> # A tibble: 3 × 6
#>   Species     mean    sd     n     se     ci
#>   <fct>      <dbl> <dbl> <int>  <dbl>  <dbl>
#> 1 setosa      5.01 0.352    50 0.0498 0.0977
#> 2 versicolor  5.94 0.516    50 0.0730 0.143 
#> 3 virginica   6.59 0.636    50 0.0899 0.176

By default, summariser uses a normal distribution to calculate confidence intervals. If you would rather use a t distribution, just pass this to the type parameter.

iris %>%
  group_by(Species) %>% 
  summary_stats(Sepal.Length, type = "t")

condwanaland/summariser documentation built on Sept. 4, 2021, 3:44 a.m.