Answer:
a) 95 % confidence interval for the mean estimate of the average income of all U.S. households based on the information
(58,808 ,71,192)
Step-by-step explanation
Step-by-step explanation:-
Given data assume the 25 adults in the study can be considered an SRS from the population of all adult citizens of the United States.
Given sample size 'n' =25
The mean estimate x⁻ = $ 65 , 000
sample standard deviation 's' = $ 15 , 000
Degrees of freedom γ= n-1 =25-1=24
The tabulated value t = 2.0639 (check t- table blue mark) at 0.05 level of significance with 24 degrees of freedom.
Step(ii):-
95 % confidence interval for the mean


(65,000-6191.7 ,65,000+6191.7)
(58,808 ,71,192)
Conclusion:-
95 % confidence interval for the mean estimate of the average income of all U.S. households based on the information
(58,808 ,71,192)