Problem 35, page 376: One way doctors try to determine how long a single dose of
ID: 3359719 • Letter: P
Question
Problem 35, page 376: One way doctors try to determine how long a single dose of pain-reliever will provide relief is to measure the drug's half-life, which is the length of time it takes for one half of the dose to be eliminated from the body. The following table provides the half-lives of the pain reliever oxycodone for a sample of 18 individuals: 3.3 4.7 1 2.0 2.5 3.7 4.3 5.0 1.2 6.0 2.8 3.9 3.5 2.1 4.8 3.0 4.9 a) Check for normality: Are there any outliers? b) Construct a 95% confidence interval for this sample c) The NIH report that the mean half-life for Oxycodone is 3.51, does this sample contradict this mean number? d) Would it be surprising if a sample mean from a different sample showed a mean of about 4.5 hours?Explanation / Answer
Part a
We know that if the value is below 1.5*IQR from Q1 or above 1.5*IQR from Q3, then this value considered as an outlier.
For the given variable X, we have following descriptive statistics:
Variable N Mean Median TrMean StDev SE Mean
X 18 3.583 3.600 3.581 1.359 0.320
Variable Minimum Maximum Q1 Q3
X 1.200 6.000 2.400 4.825
IQR = Q3 – Q1 = 4.825 – 2.4 = 2.425
1.5*IQR = 1.5*2.425 = 3.6375
Q1 - 1.5*IQR = 2.4 - 3.6375 = -1.2375
Q3 + 1.5*IQR = 4.825 + 3.6375 = 8.4625
All values are lies within -1.2375 and 8.4625, so there is no any outlier.
Part b
Now, we have to find 95% confidence interval.
Confidence interval = Xbar -/+ t*S/sqrt(n)
We are given
Xbar = 3.583
S = 1.359
n = 18
df = n – 1 = 18 – 1 = 17
Confidence level = 95%
Critical t value = 2.1098
(By using t-table or excel)
Confidence interval = 3.583 -/+ 2.1098*1.359/sqrt(18)
Confidence interval = 3.583 -/+ 2.1098* 0.320319372
Confidence interval = 3.583 -/+ 0.6758
Lower limit = 3.583 - 0.6758 = 2.9072
Upper limit = 3.583 + 0.6758 = 4.2588
Confidence interval = (2.9072, 4.2588)
Part c
This sample does not contradict the given mean number 3.51, because it is lies within given confidence interval.
Part d
Yes, it would be surprising if a sample mean from a different sample showed a mean of about 4.5 hours, because the value 4.5 is not lies within given confidence interval.
Related Questions
drjack9650@gmail.com
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.