Problem: By end of 2015 you received an average of 129.2 complaints (i.e. individual items to be fixed under warranty) per new car sold, with a standard deviation of 42.1 complaints based on 3834 new cars. This year you have set up a quality assurance program to fix some of these problems before a car is delivered. So far this year 2016, you have had an average of just 93.4 complaints per new car sold with a standard deviation of 37.7, based on 74 cars sold so far.

To see if your new quality assurance program is working, what hypothesis testing method would you use?

Identify the populations, samples and hypotheses.

Perform a two-tailed test at the 5% level and report the results