Questions 19-20 are based on the following. Blood banks will shortly start to screen all donors for NANB hepatitis. Although the new screening tests are estimated to disqualify up to 5 percent of all prospective blood donors, they will still miss two-thirds of donors carrying NANB hepatitis. Therefore, about 10 percent of actual donors will still supply NANB-contaminated blood. 19. The argument above depends on which of the following assumptions? (A) Donors carrying NANB hepatitis do not, in a large percentage of cases, carry other infections for which reliable screening tests are routinely performed. (B) Donors carrying NANB hepatitis do not, in a large percentage of cases, develop the disease themselves at any point. (C) The estimate of the number of donors who would be disqualified by tests for NANB hepatitis is an underestimate. (D) The incidence of NANB hepatitis is lower among the potential blood donors than it is in the population at large. (E) The donors who will still supply NANB-contaminated blood will donate blood at the average frequency for all donors.

引用战友的分析：数字题，我觉得，一定要想清楚是怎样推出结论的，即：先把数学推导部分搞定。另外：数字是不可以怀疑的。 19. 数学推导可得，about 10 percent of actual donors will still supply NANB-contaminated blood. 答案A表明：有病捐血者不会因为其他原因而被排除在外。即：没有其他可能使这个10%降低，当然就是原题的assumption了。 C underestimated 竟然怀疑数学推导，斯拉斯拉迪油。

题目中的数字是不可怀疑的？从题目给的那几个数字5%，2/3，就直接说10%？我总感觉这道题有些jump into the 10%，有问题，它还处在逻辑题里，感觉要接受这种思维有些困难。敢问各路高手，是不是这题如果是weaken的话，是不是可以从这点大大的weaken一下，说这是它的inference flaw?

但是这道题不从这里问，而是让你接受这么一个推理之后再寻找它的前提，这时是不是就要无条件接受这种明显的让人逻辑思维不能正常进行的怪条件？这也是一个原则？做这样的题都不要怀疑题目本身？