Perform multiple paired t-tests based on groups/categories

The tidy way of doing it is using dplyr and broom:

library(dplyr)
library(broom)

df <- data %>% 
  group_by(Product_type) %>% 
  do(tidy(t.test(.$Price_Online, 
                 .$Price_Offline, 
                 mu = 0, 
                 alt = "two.sided", 
                 paired = TRUE, 
                 conf.level = 0.99))))

Much more readable than my base r solution, and it handles the column names for you!

EDIT
A more idiomatic way to do it rather than using do (see r4ds) is to use nest to create nested dataframes for each product type, then run a t-test for each nested dataframe using map from purrr.

library(broom)
library(dplyr)
library(purrr)
library(tidyr)

t_test <- function(df, mu = 0, alt = "two.sided", paired = T, conf.level = .99) {
  tidy(t.test(df$Price_Offline, 
              df$Price_Online,
              mu = mu, 
              alt = alt,
              paired = paired,
              conf.level = conf.level))
}

d <- df %>%
  group_by(Product_type) %>%
  nest() %>%
  mutate(ttest = map(data, t_test)) %>%
  unnest(ttest, .drop = T)

Leave a Comment