In: Statistics and Probability
A pharmacologist decided to test two common headache-tablets for their effectiveness: Tablet A 500mg and Tablet B 20 mg. The experiment was conducted as follows: A random patient that walked into the clinic and complained of a headache was given either tablet A, tablet B, or a placebo. After swallowing the tablet the patient was asked to stay in the clinic for an hour and afterwards to report whether the headache had disappeared, improved, or if the tablet had no effect (i.e. no improvement or even a worsening of the headache intensity). Use a 1% level of significance to test the claim that headache status is independent of headache relief tablet used.
Headache Disappeared |
Headache Improved |
No Change |
Total |
|
Tablet A |
70 |
20 |
15 |
105 |
Tablet B |
60 |
10 |
20 |
90 |
Placebo |
40 |
35 |
25 |
100 |
Total |
170 |
65 |
60 |
295 |
Contingency Table: Headache 6
Hypotheses:
H0: Headache status is independent of headache
relief tablet used.
H1: Headache status is dependent on headache
relief tablet used.
Expected Values:
Complete the 3x3 table of expected outcomes (round values to 3
decimal places).
|
Headache Disappeared |
Headache Improved |
No Change |
Tablet A |
23.136 |
||
Tablet B |
18.305 |
||
Placebo |
57.627 |
Results:
Calculate the test statistic (use two decimal
places).
State the p-value (round answer to the nearest hundredth of a percent - i.e. 1.53%)
p-value =
Conclusion:
We sufficient evidence to support the claim that patient
headache status is dependent on which headache relief tablet was
used (p 0.01).
(Use “have” or “lack” for the first blank and “<” or “>” for
the second blank.)
The Expected value data are in the table below. Each Cell = Row total * Column Total/N. N = 295
The degrees of freedom, df = (r – 1) * (c -1) = (3 - 1) * (3 - 1) = 4
Expected | ||||
Disappeared | Improved | No Change | Total | |
Transportation | 60.509 | 23.136 | 21.356 | 105 |
Information | 51.864 | 19.831 | 18.305 | 90 |
Financial | 57.627 | 22.034 | 20.339 | 100 |
Total | 170 | 65 | 60 | 295 |
______________________________
The Test Statistic: The table below gives the calculation of .
# | Observed | Expected | (O-E) | (O-E)2 | (O-E)2/E |
1 | 70 | 60.509 | 9.4915 | 90.08857 | 1.4889 |
2 | 60 | 51.864 | 8.1356 | 66.18799 | 1.2762 |
3 | 40 | 57.627 | -17.6271 | 310.7147 | 5.3918 |
4 | 20 | 23.136 | -3.1356 | 9.831987 | 0.4250 |
5 | 10 | 19.831 | -9.8305 | 96.63873 | 4.8732 |
6 | 35 | 22.034 | 12.9661 | 168.1197 | 7.6300 |
7 | 15 | 21.356 | -6.3559 | 40.39746 | 1.8916 |
8 | 20 | 18.305 | 1.6949 | 2.872686 | 0.1569 |
9 | 25 | 20.339 | 4.661 | 21.72492 | 1.0681 |
Total | 24.202 |
test = 24.20
______________________
The p value: The p value at test = 24.2, df = 4; P value = 0.0001
______________________
The Conclusion: We have sufficient evidence to to support the claim that patient headache status is dependent on which headache relief tablet was used (p < 0.01)
_____________________