S4 June 2014 (R) Q1
1. In a trial for a new cough medicine, a random sample of 8 healthy patients were given steadily increasing doses of a pepper extract until they started coughing. The level of pepper that triggered the coughing was recorded. Each patient completed the trial after taking a standard cough medicine and, at a later time, after taking the new medicine. The results are given in the table below.
| Level of pepper extract that triggers coughing | ||||||||
|---|---|---|---|---|---|---|---|---|
| Patient | \(A\) | \(B\) | \(C\) | \(D\) | \(E\) | \(F\) | \(G\) | \(H\) |
| Standard medicine | 46 | 12 | 18 | 31 | 23 | 16 | 27 | 9 |
| New medicine | 53 | 16 | 13 | 49 | 11 | 34 | 38 | 22 |
| Scheme | Marks |
|---|---|
| [New − standard = ] \(d\): 7, 4, −5, 18, −12, 18, 11, 13. | M1 |
| \(\bar{d} = 6.75\) | M1 |
| \({s_d}^2 = \dfrac{1172 - 8 \times 6.75^2}{7} = 115.3571\ldots\) or \(s_d = 10.7404\ldots\) | M1 |
| \(\mathrm{H}_0 : \mu_d = 0 \qquad \mathrm{H}_1 : \mu_d \gt 0\) | B1 |
| \(t_7 = \dfrac{6.75}{s_d/\sqrt{8}} = 1.7775\ldots\) or \(\dfrac{c}{s_d/\sqrt{8}} = 1.895\ \therefore\ \text{CR } c \gt \text{awrt } 7.2\) awrt 1.78 | M1 A1 |
| \(t_7(5\%)\) one tail critical value is 1.895 (or prob. = 0.05935…) | B1 |
| Not significant. There is insufficient evidence that the new medicine is better or the new medicine is not recommended. | A1ft |
| (8) |
Notes
1st M1 for attempting the \(d\)s
2nd M1 for attempting \(\bar{d}\)
3rd M1 for attempting \(s_d\) or \({s_d}^2\)
1st B1 for both hypotheses correct in terms of \(\mu\) or \(\mu_d\)
4th M1 for attempting the correct test statistic \(\dfrac{6.75}{s_d/\sqrt{8}}\) or \(p = \text{awrt } 0.06\) or \(\dfrac{c}{10.7/\sqrt{8}} = t\) value
1st A1 1.78 or awrt 0.06 or awrt 7.2
2nd B1 1.895 or awrt 0.06
2nd A1ft for a correct comment in context based on their test statistic and their cv.
| Scheme | Marks |
|---|---|
| Need the differences between levels triggering coughing to be normally distributed | B1 |
| (1) | |
| (9 marks) |
Notes
B1 for a comment that mentions “differences” and “normal” distribution