Comment by gk1

When a positive outcome is almost certain, there's less need for fine measurements. For example, how much effort is it worth to determine whether tech job placement for Howard University students goes up by 2x or exactly 2.15x?