In: Statistics and Probability
The National Prenatal Statistics Unit of the Sydney Children’s Hospital in reports that the mean birth weight of all babies born in Australia in 2011 was 7.48 pounds.
A Florida hospital reports that the average weight of 118 babies born there last year was 7.52 pounds, with a standard deviation of 1.31 pounds.
If we believe that Florida babies fairly represent American newborns, is there any evidence that U.S. babies and Australian babies do not weigh the same amount at birth?
The Hypotheses for the test are
HO : μ = 7.48
HA : μ ≠ 7.48
1. Find the P-value for the test. (Due to the different methods, round to 2 decimal places and select the best possible answer)
P-value: [ Select ] ["0.04", "0.21", "0.03", "0.74"]
2. Using an α = 0.10 significance level and your P-value , we would conclude:
Decision:
Conclusion:
3. What would be a Type I Error in the context of this problem?
4. Find a 90% confidence interval for the true mean weight of American newborns
Solution-1:
t=xbar-mu/s/sqrt(n)
=(7.52 -7.48)/(1.31/sqrt(118))
t=0.331688
df=n-1=118-1=117
two tail p value in excel
=T.DIST.2T(0.331688,117)
=0.74071827
p=0.74
2. Using an α = 0.10 significance level and your P-value , we would conclude:
Decision:p=0.74>0.10
Conclusion:Fail to reject Ho
3. What would be a Type I Error in the context of this problem?
Type 1 error=alpha=0.10
Solution-4:
df=n-1=118-1=117
alpha=0.10
alpha/2=0.10/2=0.05
t crit in excel
==T.INV(0.05,117)
=1.657981659
90% confidence interval for mean
xbar-tc*s/sqrt(n),xbar+tc*s/sqrt(n)
7.52-1.657981659*1.31 /sqrt(118),7.52+1.657981659*1.31 /sqrt(118)
7.320055, 7.719945
we are 90% confident that the true mean weight of American newborns lies in between 7.320055 and 7.719945