Question 332023
You should set up the hypothesis such that type I errors are worse than type II errors
----
Ho: Not all public guest rooms have smoke detectors with adjustable sensitivity
Ha: All public guest rooms have smoke detectors with adjustable sensitivity
----
Type I error: 
Probability(Reject Ho / when Ho is true)
which means you conclude that all guest rooms have the detectors when in fact they all dont.  
Consequences:  You assume a level of safety that does not exist, which could lead to fire damage, injury and lawsuits.
------
Type II error: 
Probability(Accepting Ho / when Ho is false)
which means you conclude that NOT all guest rooms have the detectors when in fact they all do.  
Consequences:  You incur extra costs,by buying more detectors than you need