IN 1983, THE so-called crack epidemic swept across the United States. To some pe
ID: 3358144 • Letter: I
Question
IN 1983, THE so-called crack epidemic swept across the United States. To some people, though, the issue was nothing new. Nearly 35 years later, the U.S. is facing another widespread drug problem. Overdoses from opioids (which include heroin as well as prescription meds like oxycodone) killed more than 33,000 Americans in 2015, according to the Centers for Disease Control and Prevention. The total marked an all-time high. The New York Times estimates that the death rate from drug overdoses increased by 19 percent in 2016. It is a devastating trajectory.
What would the null hypothesis be?
What would the alternate hypothesis be.
Apply Type 1 and/or Type II errors?
Explanation / Answer
p1 - death ratio in year 2015
p2 - death ration in year 2016
Below are the null and alternate hypothesis
H0: p1 - p2 >= 0.19
H1: p1 - p2 < 0.19
Type I error would happen if we reject the true null hypothesis. This means though proportion is not actually increased by 19% but we still reject the null hypothesis.
Type II error would happen if we fail to reject the false null hypothesis. This means though proportion is actually increased by 19% but we still failed to reject null hypothesis.