Generate two-sample t-test results for multiple sub-groups

Source: R/t_test.R

This function produce two-sample t-test (two-tailed with confident interval at 0.95) results for multiple sub-groups and provides with a nice output in a table format. It can also add adjusted p values for multiple comparison issue.

t_test_two_sample(data, x, y, paired = FALSE, p_adjust = "bonferroni")

Arguments

data

A grouped data frame. It should be grouped by the intended sub-groups which you want to do the same t-test.
x

Column name of the variable which contains data values that you want to test (see t.test and details).
y

Column name of the variable which contains data values of group assignments for the test values (see t.test and details).
paired

a logical indicating whether you want a paired t-test.
p_adjust

character indicating which method should be used for adjusting multiple comparisons (see p.adjust and details). The default "bonferroni" corresponds to Bonferroni adjustment.

Value

A data.frame with the t-statistics table consisting of characters. The columns that are always present are: group variable(s), tvalue, df (degrees of freedom), p, and p_adjustmethod(s).

Examples

t_test_two_sample(color_index_two_sample, x = "color_effect", y = "group", paired = TRUE)

#> Warning: The `t_test_two_sample()` function expects a grouped data frame (i.e., from `dplyr::group_by()`). Returning statistics for the overall comparison
#> Warning: `...` must not be empty for ungrouped data frames.
#> Did you want `data = everything()`?
#> # A tibble: 1 x 4
#>   tvalue    df         p p_bonferroni
#>    <dbl> <dbl>     <dbl>        <dbl>
#> 1   4.44   231 0.0000142    0.0000142

# use bonferroni and fdr method for adjusted p values.
library(magrittr)
color_index_two_sample %>%
  t_test_two_sample(
  x = "color_effect", y = "group", paired = TRUE, p_adjust = c("bonferroni","fdr")
  )

#> Warning: The `t_test_two_sample()` function expects a grouped data frame (i.e., from `dplyr::group_by()`). Returning statistics for the overall comparison
#> Warning: `...` must not be empty for ungrouped data frames.
#> Did you want `data = everything()`?
#> # A tibble: 1 x 5
#>   tvalue    df         p p_bonferroni     p_fdr
#>    <dbl> <dbl>     <dbl>        <dbl>     <dbl>
#> 1   4.44   231 0.0000142    0.0000142 0.0000142