In: Statistics and Probability
Heart rates for newborns born in one of two hospitals were measured to determine whether birth outcomes are comparable at the two locations. Assume the hospitals are independent.
Relationship between heart rate and hospital |
|||
Hospital |
Mean heart rate (bpm) |
SD |
n |
Riverside Methodist |
127 |
11 |
216 |
NRV General |
133 |
12 |
153 |
a) Test for a significant difference between mean heart rate in the two hospitals. Remember, as always, to give hypotheses, a p-value, and any other necessary information. Choose an appropriate p-value.
We compare the length of stay in days from several randomly selected patients:
Length of stay comparison |
|
Hospital |
Length of stays (days) |
Riverside Methodist |
21, 10, 32, 60, 7, 44, 27, 3, 16, 26, 33 |
NRV General |
86, 27, 9, 70, 88, 75, 120, 60, 35, 73, 96, 44, 240 |
a) Why is a t-test inappropriate in this case?
b) Conduct an appropriate nonparametric test to determine whether the two hospitals are com- parable in length of stay. Report a conclusion and necessary work.
In first case we use two sample t test for summary data, using
MINITAB
Using minitab commands:
The command is Stat>>>Basic Statistics >>2 sample t...
Summarized data: Choose if you have summary values for the sample size, mean, and standard deviation for each sample.
First
Sample size: Enter the value for the sample size.
Mean: Enter the value for the mean.
Standard deviation: Enter the value for the standard deviation.
Second
Sample size: Enter the value for the sample size.
Mean: Enter the value for the mean.
Standard deviation: Enter the value for the standard deviation.
then click on Option select level of confidence = 1 - alpha = (1 - 0.05)*100 = 95.0
Alternative " "
Click on Ok
Again "click on OK"
We get the following output
MTB > TwoT 216 127 11 153 133 12;
SUBC> Pooled.
Two-Sample T-Test and CI
Sample N Mean StDev SE Mean
1 216 127.0
11.0 0.75
2
153 133.0
12.0 0.97
Difference = mu (1) - mu (2)
Estimate for difference: -6.00
95% CI for difference: (-8.37, -3.63)
T-Test of difference = 0 (vs not =): T-Value = -4.97 P-Value =
0.000 DF = 367
Both use Pooled StDev = 11.4248
Decision rule: 1) If p-value < level of significance (alpha)
then we reject null hypothesis
2) If p-value > level of significance (alpha) then we fail to reject null hypothesis.
Here p value = 0.00 < 0.05 so we used first rule.
That is we reject null hypothesis .
Conclusion: At 5% level of significance there are sufficient evidence to say that the significant difference between mean heart rate in two hospitals.