o**h
发帖数: 55 | 1 A 1-year study of a new drug to treat hypertension is conducted. One hundred
patients with hypertension are enrolled; 50 patients are given the new drug
and another 50 patients are given hydrochlorothiazide. All patients
completed the trial. One noted unexpected effect is increased growth of
scalp hair which occurred in those taking the new drug, a nonstatistically
significant difference (p>0.10). This effect has also been reported in
studies of other similar drugs in the new therapeutic class. |
M*****1
发帖数: 371 | |
d****y
发帖数: 2180 | 3 D Sample Size
With only 100 patients, you may not get statistical significance if the
difference between new drug and the control dug is not big. |
c***y
发帖数: 811 | 4 D
when p>0.10,u fail to reject null hypothesis,but you have chance to make
type II error(beta). To avoid of this error, you have to increase power(1-
beta). To increase power, you need increase sample size. |
o**h
发帖数: 55 | 5 有人是这样解释，你觉得合理吗？
The question is specifically asking the validity of the results, when
referring to validity it means whether there is a systematic error in the
study, in other words a Bias. Before determining an adequate sample size,
one must do a power analysis first prior starting.
Questions states that there is a non-statistically difference for the hair
growth (p=0.10), meaning that there is no difference between the control
group and the new drug group. |
o**h
发帖数: 55 | 6 so A should be the right one. |
d****y
发帖数: 2180 | 7 哈哈，这是kaplan的解释吗？好像有点印象，比较抠字眼的。 |
d****y
发帖数: 2180 | 8 个人觉得，如果是问 validity of the results，这个解释还可以接受。这里是问
validity of the conclusion，而这个conclusion是基于p=0.10是真实的推出的。 这
个P=0.10, 就像cqfly说的，可能是type II error,所以还是D合理一点。
这个study, 所有的results可能都是valid的，只不过conclusion可能会有点问题，因为样本小不能排除type II error。实际的clinical trials很多这种情况的。
如果validity of the results有问题, 那么这个study的结果完全就没法trust,是更严重的问题。 |
