Critical region
It seems as though the statistic—a difference in promotion rates of 0.2917—is on the extreme end of the permutation distribution. That is, there are very few permuted differences which are as extreme as the observed difference.
To quantify the extreme permuted (null) differences, we use the quantile() function.
The dplyr package has been loaded for you.
This exercise is part of the course
Foundations of Inference in R
Exercise instructions
Using the permuted differences, disc_perm, find the quantile values of stat such that:
- 10% of the permuted differences are above the value (0.9 quantile)
- 5% of the permuted differences are above the value (0.95 quantile)
- 1% of the permuted differences are above the value (0.99 quantile)
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
disc_perm %>%
summarize(
# Find the 0.9 quantile of diff_perm's stat
q.90 = ___(___, p = ___),
# ... and the 0.95 quantile
q.95 = ___,
# ... and the 0.99 quantile
q.99 = ___
)