|
统计问题（25）：抽样调查大样本会导致什么情况发生？ Question If a very large random sample is taken in a population survey measuring children’s height, which if any of the following statements are true? a) The mean is likely to increase b) The range is likely to increase c) The standard error of the mean is likely to increase d) The distribution will become more normal Answer Because children’s height is distributed in a nearly normal manner very large samples will probably include a few more very tall children and very short children than smaller samples and these children’s height will be roughly the same distance away from the mean. Consequently, the range (between the tallest and the shortest) will increase but the mean will remain the same. A very large sample size will reduce the uncertainty in estimating the mean, which is what the standard error of the mean is measuring. Larger sample sizes will more closely reflect the distribution of the underlying population. If this itself is skewed in some way, as income distribution is—for example, a larger sample will still not be more normally distributed. In some circumstances a larger sample size can lead to an increase in the sample mean and to an increase in the standard deviation—for example, if the distribution has some very large outlying values that might be missed by a small sample but are likely to be captured by a larger one. This phenomenon might be seen for income distribution in some geographical areas if a few millionaires lived in the area but most of the other residents were poor. 中文解释： 由于儿童的身高分布几乎是正态的，因此，与较小的样本相比，非常大的样本可能会包含更多的非常高的儿童女和非常矮的儿童，并且这些儿童的身高与均值的距离大致相同。因此，范围（最高和最短之间，即为极差）将增加，但均值将保持不变。 很大的样本量将减少估计平均值的不确定性，这个不确定性由平均值的标准误差来测量。较大的样本量将更紧密地反映总人口的分布。如果总人口分布仍然是偏态的（例如收入分配），较大的样本仍不会是正态分布。 在某些情况下，较大的样本量可能导致样本均值的增加和标准偏差的增加，例如，如果分布中的分布值非常大，则少量样本可能会忽略掉这些值，但很可能会被更大的捕获。如果一些百万富翁居住在该地区，但其他大多数居民都是穷人，那么在某些地理区域的收入分配中可能会看到这种现象。 所以答案是选择 B 更多统计问题（最近10期） 23.考虑对无症状患者进行筛查以发现疾病时，应满足以下哪个条件？
|
|
|
