In: Statistics and Probability
An herbal medicine is tested on 10 patients with sleeping disorders. The table below shows the hours of sleep patients got during one night without using the
herbal medicine and the hours of sleep the same patients got on another night after the herbal medicine was administered. Assume underlying normal distributions. Remember the applets will only accept one word labels. Data must be organized in columns.
Patient 1 2 3 4 5 6 7 8 9 10
Hours
without 1.0 1.4 3.4 3.7 5.1 5.1 5.2 5.3 5.5 5.8
Hours
with 2.9 3.3 3.5 4.4 5.0 5.0 5.2 5.3 6.0 6.5
a. At a= 0.01, does the herbal medicine increase the number of hours of sleep during one night?
b. Construct a 95% confidence interval for the mean difference.
c. What is the point estimate for this interval?
d. What is the margin of error for this interval?
e. Interpret the confidence interval in the context of this problem.
a ) let us consider d = sleep hours without - sleep hours with
the null and alternative hypothesis is
Ho:ud= 0
Ha:ud<0
using excel>Addin>phstat>two sample test
we have
Paired t Test | |
Data | |
Hypothesized Mean Difference | 0 |
Level of significance | 0.05 |
Intermediate Calculations | |
Sample Size | 10 |
DBar | -0.5600 |
Degrees of Freedom | 9 |
SD | 0.7706 |
Standard Error | 0.2437 |
t Test Statistic | -2.2981 |
Lower-Tail Test | |
Lower Critical Value | -1.8331 |
p-Value | 0.0236 |
Reject the null hypothesis |
since p value of test stat is 0.0236 >0.01 so we do not have sufficient evidence to say that the herbal medicine increase the number of hours of sleep during one night
b ) t at 95% with 9 df is 2.262
95% confidence interval = -0.56 +/- 2.262*0.2437 = (-1.111,-0.009)
c. the point estimate for this interval is -0.56
d. the margin of error for this interval is 2.262*0.2437 = 0.551
e. we are 95% confident that the difference in sleep hours for with or without medicine lies in between (-1.111,-0.009)