This is a problem with sample proportions. The proportion of patients who wait more than one hours drops from 3/10 to 2/20= 1/10 and your sample has size 20. That can be approximated by the standard normal distribution with [tex]z= \frac{0.1- 0.3}{\sqrt{\frac{0.3(1- 0.3)}{200}}}[/tex]

Question:
Records kept at a hospital show that 3 out of every 10 patients who come to A&E have to wait more than an hour before seeing a doctor.
The hospital decides to increase staffing in the A&E to reduce the proportion.
It was found that out of the next 20 patients 2 had to wait more than an hour.
Test at the 5% level of significance whether the new staffing has decreased the number of patients waiting over an hour.