Two-sided critical region

For the discrimination data, the question at hand is whether or not women were promoted less often than men. However, there are often scenarios where the research question centers around a difference without directionality.

For example, you might be interested in whether the rate of promotion for men and women is different. In that case, a difference in proportions of -0.29 is just as "extreme" as a difference of positive 0.29.

If you had seen that women were promoted more often, what would the other side of the distribution of permuted differences look like? That is, what are the smallest (negative) values of the distribution of permuted differences?

This exercise is part of the course

Foundations of Inference in R

View Course

Exercise instructions

Using the permuted differences, disc_perm, find the quantile values such that:

1% of the permuted differences are below the value (0.01 quantile)
5% of the permuted differences are below the value (0.05 quantile)
10% of the permuted differences are below the value (0.1 quantile)

Hands-on interactive exercise

Have a go at this exercise by completing this sample code.

# Use disc_perm
___ %>% 
  # ... to calculate summary stats
  ___(
    # Find the 0.01 quantile of stat
    q.01 = ___,
    # ... and 0.05
    q.05 = ___,
    # ... and 0.1 
    q.10 = ___
  )

Edit and Run Code